Peer Revenue Returns subject to peer review

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science Computer Vision

cs.CV Watch Computer Vision

Monday 29 June

Computer Vision

Unveiling Transferability in Trajectory Prediction via Latent Scene Embeddings

Theodor Westny, David Axelsson, Björn Olofsson et al. (+1 more)

The growing availability of trajectory datasets has fueled major advances in data-driven motion prediction. Yet, models trained on one dataset often fail to generalize beyond their training domain as a result of differences in scene layouts, agent behaviors, and sensing conditions. A framework that learns latent representations of datasets and quantifies their similarity using distributional metrics is presented. This large-scale study covers 24 major datasets, including the most widely used motion-prediction benchmarks, and shows that the resulting transferability scores strongly correlate with cross-dataset model performance. The results provide practical guidance for dataset selection, pretraining, and large-scale foundation models for motion prediction, paving the way toward more generalizable and robust predictive systems.
Read the full abstract

All authors Theodor Westny, David Axelsson, Björn Olofsson, Erik Frisk

Computer Vision

Streaming Gaussian Encoding for 4D Panoptic Occupancy Tracking

Maximilian Luz, Thomas Nürnberg, Yakov Miron et al. (+1 more)

Camera-based 4D panoptic occupancy tracking (4D-POT) is a promising paradigm for holistic scene understanding from multi-view imagery, enabling joint reasoning about geometry, semantics, and object identities across time. Recent mask-based pipelines achieve strong performance by propagating instance queries across frames. However, their underlying volumetric representations are typically recomputed at each timestep, limiting geometric temporal consistency, particularly under occlusion and for static scene elements. To address this limitation, we propose a streaming Gaussian encoder that maintains a persistent volumetric scene representation for 4D-POT. Our method models the scene as a fixed-size set of latent Gaussian queries that are propagated via ego-motion compensation and refreshed under a confidence-guided budget constraint. Crucially, we shape Gaussian opacities through depth-based supervision to serve as proxy for visibility, enabling confidence to accumulate as a temporally aggregated measure of persistent scene support. Together with a warmup-based multi-frame training strategy, this yields representation-level temporal coherence beyond decoder-only tracking. Extensive experiments on Occ3D-extended nuScenes and Waymo establish a new state-of-the-art for camera-based 4D-POT, improving tracking consistency with negligible computational overhead while remaining fully compatible with existing mask-based pipelines. We provide code and models at https://sge.cs.uni-freiburg.de.
Read the full abstract

All authors Maximilian Luz, Thomas Nürnberg, Yakov Miron, Abhinav Valada

Computer Vision · Cross-listed from cs.OS

LUMOS: A Semantic Operating-System Layer for Accessibility-Grounded AI Agents

Yogeswar Reddy Thota

Current operating systems expose interfaces optimized for human users but not for AI agents. Humans benefit from pixels, icons, windows, visual grouping, mouse movement, and keyboard shortcuts; AI agents instead need compact semantic state, grounded actions, and reliable feedback. As a result, many computer-use agents are forced to interpret screenshots, OCR output, and visual crops, introducing high token costs, visual ambiguity, latency, and coordinate uncertainty. This paper introduces LUMOS (Language Model Unified Machine-Readable Operating-System Semantics), a semantic interaction layer between AI agents and operating systems. LUMOS converts native accessibility metadata and browser UI structures into machine readable semantic blueprints with stable identifiers, roles, names, values, bounds, and action affordances. It also supports live semantic pointer grounding by querying the UI element under or near the cursor through operating-system automation APIs. An LLM then acts through an accessibility grounded observe act loop using constrained visible-UI primitives rather than application-specific scripts. LUMOS does not claim to replace visual agents; instead, it reduces dependence on screenshots when operating systems already provide semantic structure. These results suggest a path toward AI-native operating systems and machine-readable interaction layers.
Read the full abstract

Computer Vision

Open-Vocabulary and Referring Segmentation for 3D Gaussians Using 2D Detectors

Jameel Hassan, Yasiru Ranasinghe and Vishal Patel

3D Gaussian Splatting (3DGS) has emerged at the forefront of 3D scene reconstruction. Extending 3DGS with language-driven, open-vocabulary understanding has gained significant attention for real-world applications such as embodied AI. Recent methods achieve this by learning an instance feature attribute and assigning semantics by distilling high-dimensional Contrastive Language-Image Pretraining (CLIP) features directly into the scene representation. However, the instance grouping mechanisms of these methods either require a predefined number of instances or suffer from noise in their bottom-up grouping strategies. Furthermore, the reliance on CLIP restricts semantic understanding to simple noun phrases, preventing complex spatial reasoning and referential expression grounding. We present GaussDet, a method that circumvents the need for dense CLIP features by leveraging discrete, open-vocabulary 2D object detectors with referring expression capabilities. We learn instance features for individual Gaussians to decompose the scene into 3D instance groups. By rendering these groups and aggregating semantic votes from multi-view 2D detections, we generate a robust View-Aggregated Semantic Label Distribution (VASD) for each 3D instance. This view-aggregation strategy acts as a strong regularizer, attenuating spurious labels caused by low-quality instance grouping. Our approach enables a straightforward, zero-shot extension from simple language queries to complex referential grounding. Extensive evaluations across two key tasks – open-vocabulary segmentation (LeRF-OVS, ScanNet) and referring expression grounding (Ref-LeRF) – demonstrate that GaussDet achieves consistent improvements over existing methods. Most notably, we achieve a substantial 16.7% mIoU improvement in referential grounding within a strict zero-shot setting.
Read the full abstract

Computer Vision · Cross-listed from cs.RO

GROW$^2$: Grounding Which and Where for Robot Tool Use

Yuhong Deng, Yuyao Liu and David Hsu

Can the robot use a plate to cut a cake if no knife is available? Tool use greatly expands robot capabilities, but to use tools creatively beyond their intended functions, the robot faces the challenge of $\textit{open-world affordance grounding}$: select an open-category object to act as a tool and localize its specific region of action. To this end, we introduce GROW$^2$ (GROunding Which and Where), which leverages object parts as a natural abstraction to split the grounding process hierarchically into semantic and geometric levels, thus bypassing the need for data-heavy, end-to-end training. Semantically, GROW$^2$ harnesses the commonsense reasoning of Vision-Language Models (VLMs) to parse a natural-language task instruction, select a suitable object as the tool, and identify task-relevant parts on the tool and the target object. Geometrically, vision foundation models then ground the selected parts into precise 3D regions from a single RGB-D image. Experiments on established benchmarks show that GROW$^2$ outperforms state-of-the-art baselines on affordance prediction benchmarks. Further, it achieves zero-shot generalization over open-category objects and outperforms baselines in both simulated and real-world robot tool use experiments.
Read the full abstract

Computer Vision

Reweighting Framewise Attention in Video Transformers for Facial Expression Understanding

Seongro Yoon, Donghyeon Cho, Jinsun Park et al. (+1 more)

Understanding facial expressions in videos requires modeling subtle and localized facial dynamics under unconstrained conditions. Although recent Vision Transformer (ViT)-based video models have shown strong performance through large-scale self-supervised pretraining, their attention mechanisms often emphasize dominant global motions and coarse temporal dynamics, limiting sensitivity to fine-grained facial variations. To address this limitation, we propose MiRA (Marginal-induced Attention Redistribution), a plug-in frame-marginal attention redistribution framework for ViT backbones that enhances spatio-temporal selectivity toward subtle facial dynamics without introducing additional trainable parameters. MiRA derives frame-level confidence and intra-frame concentration statistics from self-attention maps to estimate frame-wise marginal importance and redistribute attention toward spatiotemporally localized facial cues. We first introduce a principled exact mode based on post-softmax attention redistribution. To further improve efficiency, we propose flashLite mode, a lightweight pre-softmax approximation that integrates frame-marginal redistribution into FlashAttention kernels while preserving the effectiveness of the exact formulation. Experimental results on challenging Facial Expression Recognition (FER) benchmarks demonstrate consistent improvements over strong ViT baselines.
Read the full abstract

All authors Seongro Yoon, Donghyeon Cho, Jinsun Park, François Brémond

Computer Vision

UnfoldArt: Zero-Shot Recovery of Full Articulated 3D Objects from Text or Image

Mohamed el Amine Boudjoghra, Ivan Laptev and Angela Dai

Articulated 3D objects are essential for interactive environments in embodied AI, robotics, and virtual reality, but reconstructing their structure and motion from sparse observations remains challenging. Existing approaches remain largely constrained by lack of supervised data or lack the priors needed to reliably recover articulation, hidden geometry, and internal object structure. We present the first debate-driven agentic approach to articulated 3D object reconstruction from text or image inputs that both grounds articulation reasoning in concrete motion and exposes the occluded geometry revealed under articulation. High-level agents reason about object semantics and motion using knowledge from vision-language and video models, while low-level agents estimate articulation parameters and interaction points; together, they engage in a two-round structured debate that first exploits global–local disagreement and then grounds the agents in freely generated video. The same video prior, conditioned on the agreed articulation, then drives each part through its motion to expose occluded interiors and geometry that cannot be inferred from a single static view. By combining agentic reasoning with a video generative prior, our approach jointly infers articulation and reconstructs complete 3D articulated objects, producing high-fidelity geometry, internal structure, and motion-consistent states beyond directly observed surfaces.
Read the full abstract

Computer Vision

Goku: A Million-Scale Universal Dataset and Benchmark for Instruction-Based Video Editing

Sen Liang, Cong Wang, Zhentao Yu et al. (+8 more)

Existing instruction-based video editing datasets commonly focus on single-task appearance editing, failing to meet the complex creative demands of real-world scenarios. To bridge this gap, we present Goku, a large-scale dataset featuring 2 million high-quality, instruction-aligned video editing pairs, which is the first to extend task boundaries from basic appearance editing to multi-task and structural manipulations(e.g., precise control of subject movement). To tackle the data synthesis challenges inherent in these complex tasks, we design an efficient data synthesis pipeline that decomposes complex edits into controllable sub-problems and introduce a progressive filtering system for data reliability throughout the whole process. Furthermore, we explore the optimal network structures on Goku, and propose Goku-Edit. To deeply comprehend complex editing instructions, Goku-Edit leverages an MLLM as its text encoder and adopts a decoupled dual-branch design: a dedicated mask branch handles structural control, freeing the main branch for appearance rendering. A comprehensive video editing benchmark, Goku-Bench, is also proposed with 1,000 human-verified test cases and 7 novel editing-specific metrics. Evaluated on Goku-Bench, Goku-Edit obtains up to +8% improvement on other open-source models in terms of instruction following.
Read the full abstract

All authors Sen Liang, Cong Wang, Zhentao Yu, Fengbin Guan, Zhengguang Zhou, Teng Hu, Youliang Zhang, Yuan Zhou, Xin Li, Qinglin Lu, Zhibo Chen

Computer Vision

Towards in-the-wild Egocentric 3D Hand-Object Pose Estimation

Siddhant Bansal, Zhifan Zhu, Shashank Tripathi et al. (+3 more)

Estimating accurate 3D hand-object pose from in-the-wild egocentric RGB remains challenging due to severe occlusions and ambiguous contact. Existing learning-based methods often struggle to generalise to in-the-wild scenes and are limited by the scarcity of supervision. We address these issues with two contributions. First, we introduce EPIC-Contact, an in-the-wild egocentric dataset of 2.3K clips (62.3K frames) with dense, bijective 3D hand-object contact correspondences and posed meshes. Second, we propose HOPformer, an end-to-end transformer that jointly predicts bi-manual hand and object pose in a single forward pass. A cross-attention decoder conditions object features on hand priors, producing robust pose estimation. We test HOPformer on the in-lab 3D dataset, ARCTIC, as well as our newly introduced EPIC-Contact dataset. HOPformer reaches 82.4% success rate on ARCTIC (+6.2 pts over current SOTA). On EPIC-Contact, it nearly doubles the success rate while reducing contact deviation by 75%. EPIC-Contact, HOPformer code and checkpoints are released: https://sid2697.github.io/epic-contact.
Read the full abstract

All authors Siddhant Bansal, Zhifan Zhu, Shashank Tripathi, Jiahe Zhao, Michael J. Black, Dima Damen

Computer Vision

Learning from Reliable Latent Prompts for Visual Recognition with Missing Modalities

Taixi Chen and Nancy Guo

Large-scale multimodal models (LMMs) have achieved superior performance in visual recognition by synergizing information across diverse, massive-scale paired modalities. In real-world scenarios, however, missing-modality inputs are ubiquitous, causing models optimized for modality-complete data to exhibit precipitous performance degradation. Existing research has introduced prompt learning to mitigate this issue, typically by generating dynamic prompts from instance-level features, regardless of whether the input modalities are complete or partially absent. However, such input-conditioned strategies are hindered by the escalating unreliability of instance-level features; as higher missing rates increase the proportion of incomplete modalities, the resulting instability in prompt learning limits the model's performance. To address this limitation, we hypothesize that learnable latent prompts themselves encapsulate stable, modality-intrinsic priors that are decoupled from corrupted inputs. Consequently, we propose a novel paradigm: Learning from Reliable Latent Prompts. Unlike prior methods, we model input-agnostic learnable prompts as stable latent anchors that enable robust guidance and effective cross-modal knowledge compensation, even under extreme missing rates (e.g., 90%). Empirical results across three benchmark datasets demonstrate that our "learn-from-latent-prompts" approach achieves state-of-the-art performance across a wide range of missing-modality scenarios. Extensive experiments further confirm the effectiveness of this paradigm in providing a robust solution to the missing-modality problem.
Read the full abstract

Computer Vision

APRIL-MedSeg: A Modular Medical Image Segmentation Toolbox Embracing Modern Paradigms

Juntao Jiang, Jinsheng Bai, Linxuan Fan et al. (+3 more)

We present APRIL-MedSeg, a YAML-driven modular framework for 2D medical image segmentation. It provides a unified and extensible ecosystem that decomposes segmentation networks into reusable components. Also, the framework integrates a broad spectrum of advanced paradigms, including semi-supervised learning, domain adaptation, knowledge distillation, weakly supervised learning, and text-guided segmentation as well as foundation model support. A registry-based configuration system with inheritance enables flexible and reproducible experiment management, supporting seamless switching across models, datasets, and training strategies. In addition, the framework provides a unified interface for medical datasets, augmentation pipelines, deployment utilities and model ensembling. Overall, APRIL-MedSeg is designed as a general-purpose research and development platform that bridges algorithmic innovation and practical deployment, while also serving as a structured ecosystem for systematically organizing and reproducing advances in medical image segmentation. The code is available at https://github.com/juntaoJianggavin/APRIL-MedSeg under an Apache 2.0 license.
Read the full abstract

All authors Juntao Jiang, Jinsheng Bai, Linxuan Fan, Yali Bi, Jiangning Zhang, Yong Liu

Computer Vision

Beyond 2D Matching: A Unified Single-Stage Framework for Geometry-Aware Cross-View Object Geo-Localization

Liyao Wang, Ruipu Wu, Haojun Xu et al. (+3 more)

Cross-view object geo-localization (CVOGL) aims to locate a target object from a query view (e.g., ground or drone) within a geo-tagged reference image (e.g., satellite). Existing approaches heavily rely on 2D appearance matching and are constrained by limited datasets lacking geometric metadata, diverse prompts, and standard field-of-view imagery. To address these intertwined challenges, we first introduce \dataset, a large-scale, high-fidelity building dataset comprising over 220,000 ground-satellite and drone-satellite pairs. It provides multi-modal prompts (points, boxes, masks) and camera poses to enable flexible target referring and explicit spatial modeling. Furthermore, we propose a novel single-stage Geometry-Aware Geo-localization framework (GAGeo), built upon the permutation-equivariant 3D foundation model $\pi^3$. By seamlessly integrating visual features, referring prompts, and learnable task tokens, our model adapts the inherited 3D prior to jointly predict bounding boxes, segmentation masks, and camera poses in a single forward pass. Additionally, we introduce a contrastive loss that utilizes the satellite view as a universal anchor, implicitly aligning ground and drone representations to enable zero-shot ground-to-drone localization without requiring triplet training data. Extensive experiments demonstrate that our approach significantly outperforms state-of-the-art methods, exhibiting exceptional generalization ability in unseen scenes and novel cross-view setups.
Read the full abstract

All authors Liyao Wang, Ruipu Wu, Haojun Xu, Lei Shi, Linjiang Huang, Si Liu

Computer Vision · Cross-listed from cs.AI

The Human Creativity Benchmark

Aspen Hopkins, Allison Nulty, Alexandria Minetti et al. (+2 more)

Modern AI evaluation frameworks treat evaluator disagreement as noise to be resolved. In creative domains, professional disagreement reflects genuine differences in taste, not measurement error. We argue that evaluating creative AI requires preserving two distinct signals: convergence, where professionals align around shared best practices, and divergence, where individual taste legitimately varies. We present the Human Creativity Benchmark (HCB), a benchmark that operationalizes this separation by collecting pairwise preferences, scalar ratings on prompt adherence, usability, and visual appeal, and qualitative rationale from domain professionals. Across 15,000 professional judgments spanning five creative domains and three workflow phases (ideation, mockup, refinement), we find that convergence concentrates on verifiable dimensions like technical correctness and visual hierarchy, while divergence concentrates on taste-driven dimensions like aesthetic direction and conceptual risk. No model excels uniformly across all phases. Collapsing these signals into a single quality metric discards the most actionable information: where models must be correct versus where they should remain steerable.
Read the full abstract

All authors Aspen Hopkins, Allison Nulty, Alexandria Minetti, Anoop Pakki, Angad Singh

Computer Vision

EcoVideo: Entropy-Orchestrated Video Generation Paradigm in Cloud-Edge Dynamics

Jiayu Chen, Hengyi Zhang, Maoliang Li et al. (+5 more)

DiT video generation is latency-intensive due to iterative full-frame denoising, while prior cloud-edge methods largely rely on static inter-step decoupling and cannot leverage inter-frame similarity or adapt to system dynamics. We propose EcoVideo, an entropy-orchestrated framework for dynamic inter-frame decoupling: early-stage self-attention entropy provides a training-free estimate of frame-wise information density for frame selection; a cloud large model denoises sparse high-entropy keyframes; and an edge lightweight model reconstructs the remaining frames via motion-aware interpolation with refinement for temporal stability. EcoVideo further adapts the keyframe budget and edge refinement depth to real-time bandwidth and compute availability, optimizing end-to-end latency under constraints. Experiments on representative DiT video generators show improved quality–efficiency trade-offs and up to 2.9x end-to-end speedup in low-bandwidth, compute-limited edge settings. Code is available at https://github.com/IF-LAB-PKU/EcoVideo.
Read the full abstract

All authors Jiayu Chen, Hengyi Zhang, Maoliang Li, Minyu Li, Zihao Zheng, Xuanzhe Liu, Guojie Luo, Xiang Chen

Computer Vision · Cross-listed from cs.RO

Training Vision-Language-Action Models with Dense Embodied Chain-of-Thought Supervision

Haoyang Li, Guanlin Li, Youhe Feng et al. (+9 more)

Cross-embodiment transfer in vision-language-action (VLA) models remains challenging because low-level state and action spaces differ fundamentally across robot platforms. We observe that the high-level cognitive process underlying manipulation, including scene perception, object identification, task planning, and sub-task decomposition, is largely shared across embodiments. Based on this observation, we present ZR-0, a 2.6 billion parameter end-to-end VLA model that uses dense Embodied Chain-of-Thought (ECoT) supervision to align cross-embodiment representations within the vision-language model (VLM). ZR-0 adopts a dual-stream architecture: a pre-trained VLM (System 2) generates structured ECoT reasoning during training, while a Diffusion Transformer-based action expert (System 1) produces continuous action chunks via flow matching. The two components are coupled through cross-attention, with an attention mask that restricts the action expert to input prompt features only, enabling ECoT generation to be entirely skipped at inference without any performance loss. ZR-0 is pre-trained on ProcCorpus-60M, a large-scale dataset comprising approximately 60 million frames (approximately 1,000 hours) from over 400K trajectories, with dense ECoT annotations covering 96.8% of all frames. We evaluate ZR-0 on three simulation benchmarks spanning single-arm (LIBERO), bimanual (RoboTwin 2.0), and humanoid (RoboCasa GR-1 Tabletop) embodiments, as well as real-world experiments on the xArm platform, demonstrating strong performance across all settings. Code and model checkpoints are available at https://github.com/RUCKBReasoning/ZR-0.
Read the full abstract

All authors Haoyang Li, Guanlin Li, Youhe Feng, Chen Zhao, Zhuoran Wang, Yang Li, Qizhe Wei, Shifeng Bao, Haitao Shen, Yihan Zhao, Tong Yang, Jing Zhang

Computer Vision

StereoGS: Sparse-View 3D Gaussian Splatting via Stereo Priors

Wenhao Yuan, Yiyuan Ge and Deli Cai

3D Gaussian Splatting (3DGS) has achieved remarkable success in real-time novel view synthesis, yet it suffers from severe overfitting under sparse-view settings due to insufficient geometric constraints. While recent methods introduce monocular depth priors to mitigate this, they inherently struggle with scale ambiguity and cross-view inconsistency, leading to defective geometry. In this paper, we propose StereoGS, a novel sparse-view 3DGS framework that integrates stereo priors to establish reliable binocular consistency. Unlike scale-agnostic monocular constraints, StereoGS introduces a Stereo Depth Regularization by constructing virtual stereo pairs during optimization and leveraging a foundation stereo model to enforce absolute scale and binocular-consistent structures. To further suppress overfitting and eliminate redundant primitives, we design a Gradient-Aware Opacity Decay strategy that dynamically penalizes Gaussians based on their relative opacity gradient magnitudes. Combined with a Consistency-Aware Dense Initialization using zero-shot multi-view depth estimation, StereoGS effectively anchors primitives to accurate scene surfaces. Extensive experiments on LLFF, DTU, Mip-NeRF360, and Blender datasets demonstrate that StereoGS achieves state-of-the-art performance in sparse-view settings without incurring any additional inference overhead. Project Page: https://stringerywh00.github.io/StereoGS_project_page/
Read the full abstract

Computer Vision · Cross-listed from cs.RO

Learning from Mistakes: Rollout-Retrieval Lifelong Policy Learning for Autonomous Driving

Cheng Gong, Haoyang Wang, Chao Lu et al. (+2 more)

Autonomous driving policies should be able to improve continually as deployment exposes them to increasingly diverse and long-tail traffic situations. However, most learning-based policies are trained or fine-tuned on expert demonstrations and then rely largely on generalization to handle challenging closed-loop scenarios, lacking an explicit mechanism to correct and retain the mistakes exposed in these scenarios. This paper studies autonomous driving policy improvement from a lifelong learning perspective: Can a pretrained policy improve continually by accumulating corrective knowledge derived from its own mistakes, while retaining previously acquired driving competence? To answer this question, we propose Rollout-Retrieval Lifelong Policy Learning (R$^2$LPL), a policy learning framework that retrieves corrective targets from recoverable policy-induced mistakes and retains the resulting knowledge through lifelong policy learning. R^2LPL addresses a key bottleneck in continual policy improvement: closed-loop mistakes reveal where the policy is weak, but do not directly specify what the policy should learn. By filtering recoverable mistake-related states and retrieving feasible corrective targets, R$^2$LPL turns sparse failure evidence into compact supervised knowledge for stable and sample-efficient policy improvement. We evaluate R$^2$LPL on large-scale closed-loop nuPlan benchmarks. With only a few rollout and continual-learning cycles, R$^2$LPL elevates a learning-based planner with moderate initial performance to state-of-the-art performance across the evaluated benchmarks, especially on the challenging and long-tail Test14-hard split. These results demonstrate the effectiveness of R$^2$LPL in converting recoverable closed-loop mistakes into corrective knowledge for sustained policy improvement.
Read the full abstract

All authors Cheng Gong, Haoyang Wang, Chao Lu, Zirui Li, Jianwei Gong

Computer Vision

Orca: The World is in Your Mind

Yihao Wang, Yuheng Ji, Mingyu Cao et al. (+54 more)

We introduce Orca, an initial instantiation of a general world foundation model. Orca learns a unified world latent space from multimodal world signals and exposes it through multimodal readout interfaces. Rather than optimizing isolated next-token, next-frame, or next-action prediction, we are centered on Next-State-Prediction modeling, offering a unified state-transition modeling route toward understanding, predicting, and acting upon the world. Orca learns through two complementary paradigms: unconscious learning captures dense natural state transitions from continuous videos, and conscious learning models sparse meaningful state transitions by language-described events and VQA supervision. For pre-training, we construct a large-scale world-learning inventory data, including 125K hours of video data and 160M event annotations. After pre-training, Orca learns a unified world latent space. To examine whether the learned latent supports downstream, we evaluate it by three representative downstream readouts: text generation, image prediction, and embodied action generation. Orca's backbone is frozen, and only the lightweight modality-specific decoders are trainable. Experiments show the scalability of the proposed paradigm and verify that stronger world latent enables stronger downstream readouts. Orca outperforms similar-sized specialized baselines. These results show that Orca, as a general world foundation model, presents a promising approach to understanding, predicting, and acting upon the world. Finally, we discuss the current limitations, aiming to provide useful insights and inspiration for the community.
Read the full abstract

All authors Yihao Wang, Yuheng Ji, Mingyu Cao, Yanqing Shen, Runze Xiao, Huaihai Lyu, Senwei Xie, Euan Liu, Klara Tian, Tianfeng Long, Yichi Zhang, Zhengliang Cai, Ruike Chen, Jifan Zhao, Ruochuan Shi, Zihan Tang, Jing Lyu, Wenxing Tan, Ningbo Zhang, Yangtao Hu, Yuming Gao, Xiansheng Chen, Junkai Zhao, Congsheng Xu, Boan Zhu, Ziqi Wang, Yupu Feng, Qiongqiong Zhang, Yingli Zhao, Yulong Ao, Shaoxuan Xie, You Liu, Guocai Yao, Leiduo Zhang, Xiaodan Liu, Yunyan Zhang, Yance Jiao, Xinyan Yang, Jiaxing Wei, Xu Liu, Tengfei Pan, Shaokai Nie, Chunlei Men, Sen Cui, Xiaojie Jin, Hongyang Li, Jianlan Luo, Yao Mu, Yunchao Wei, Jun Yan, Hang Zhao, Xiaolong Zheng, Jiaming Li, Yonghua Lin, Tiejun Huang, Zhongyuan Wang, Pengwei Wang

Computer Vision

$\mu$Flow: Leveraging Average Images for Improving Generalisation of Deepfake Faces Detectors

Orazio Pontorno, Mattia Litrico, Luca Guarnera et al. (+2 more)

Current generative models, including GANs and diffusion models, have reached an outstanding level of photorealism, posing significant risks to privacy and security. To ensure real-world applicability, deepfake detectors must generalise effectively to unseen generators. However, most existing approaches rely on supervised training with both real and fake images, which limits their generalisation especially across generators categories (e.g. GANs vs DMs). In this work, we introduce $\mu$Flow, a one-class deepfake detector trained only on real images without relying on pseudo-deepfakes or synthetic artifacts. Our approach builds on the observation that averaging multiple images amplifies consistent generative traces, producing highly discriminative feature representations. We leverage this property by modelling the distribution of features extracted from averaged images and training a normalizing flow to align the feature space of individual images with this distribution. This alignment yields a likelihood-based criterion that separates real and fake samples while promoting strong generalisation. We evaluate $\mu$Flow on a fully out-of-distribution setting, where both real and fake datasets are unseen during training. Experimental results show that our method significantly outperforms SOTA detectors. Project page: https://opontorno.github.io/MuFlow.
Read the full abstract

All authors Orazio Pontorno, Mattia Litrico, Luca Guarnera, Mario Valerio Giuffrida, Sebastiano Battiato

Computer Vision

HASTE: A Framework for Training-Free, Dynamic, and Steerable Compression of Pre-Trained Convolutional Neural Networks

Lukas Meiner, Jens Mehnert and Alexandru Paul Condurache

Deploying large convolutional neural networks (CNNs) on resource-constrained devices is challenging due to their high computational cost. While dynamic execution methods are promising, existing approaches for CNNs typically require specialized training or fine-tuning, limiting their effectiveness when applied to pre-trained models and requiring data access. To address this gap, we propose HASTE (Hashing for Tractable Efficiency), a plug-and-play convolution module that enables training-free, dynamic compression of large pre-trained CNNs. At inference time, HASTE uses locality-sensitive hashing to identify and merge redundant channels of latent feature maps on a patch-wise basis. This process simultaneously compresses the depth of both input features and their corresponding filters, resulting in computationally cheaper convolutions. We conduct extensive experiments on CIFAR-10 and ImageNet across a range of architectures, demonstrating a 46.2% FLOPs reduction in a ResNet34 on CIFAR-10 with only a 1.25% drop in accuracy, without any retraining. We support our claims by comprehensive ablation studies to validate our core design choices, an analysis of the method's properties and limitations, and a discussion that connects our channel merging scheme to the conceptually related task of token merging in Vision Transformers. Our results demonstrate that HASTE provides an effective solution for steerable compression of pre-trained CNNs at runtime, opening new possibilities for the deployment of efficient deep learning methods.
Read the full abstract

Computer Vision

3D Scene-Adaptive Trajectory-Controllable Human Image Animation with Camera Movement

Deyin Liu, Jicheng Xu, Lin Yuanbo Wu et al. (+4 more)

Human image animation, which aims to generate a video of a reference subject following a provided action sequence, has received increasing research interest. With the development of diffusion-based/flow-based video foundation models, existing animation works have began to upgrade the guidance information from 2D skeleton/pose to 3D modeling conditions. Despite achieving reasonable results, these approaches face challenges in synthesizing trajectory-controllable human motion within natural scene under changed camera views. In this work, we present a scene-adaptive human image animation framework that controls both human motion and camera trajectories within a reconstructed 3D environment for video generation. To achieve this, we first develop a ground-adaptive 3D motion retargeting approach to enable user-friendly motion trajectory control adapting to the changes of elevations of ground and orientations automatically. Then we design a viewpoint-adaptive latent fusion mechanism to inject point-cloud geometric priors through scene-visibility masking into the generative process, providing precise guidance of viewpoint changes under camera control. Experiments on two standard human image animation benchmark datasets demonstrate remarkable improvements of our method over the state of the arts in related video generation metics. Project page: https://robinhood256100.github.io/web-disp
Read the full abstract

All authors Deyin Liu, Jicheng Xu, Lin Yuanbo Wu, Xiaowei Zhao, Xiatian Zhu, Zhe Jin, Anjan Dutta

Computer Vision

High-Resolution Flood Mapping With Sentinel-1 and Sentinel-2 via Misalignment-Robust Cross-Sensor Learning and Generative Despeckling

David Ma, Jeremy Feinstein, Shreya Pandit et al. (+2 more)

Reliable high-resolution flood extent mapping from satellite imagery remains constrained by limited data fidelity and sensor-specific artifacts. Multispectral optical imagery is degraded by clouds, shadows, and urban confounders, while synthetic aperture radar (SAR) imagery is affected by speckle noise and sensor co-registration uncertainty. This work presents an integrated flood mapping framework that jointly addresses these limitations through curated datasets and novel learning strategies. We introduce a new Sentinel-2 (S2) and Sentinel-1 (S1) dataset covering the contiguous United States, featuring pixel-accurate 10 m water masks with emphasis on challenging weather conditions and urban environments that are underrepresented in existing benchmarks. High-quality S2 annotations are manually produced using rigorous geospatial labeling protocols and transferred to SAR imagery through weakly labeled temporally coincident acquisitions. To address SAR-specific artifacts, a shift-invariant loss function is employed to tolerate residual geolocation uncertainty between SAR imagery and optical-derived labels, and a Conditional Variational Autoencoder (CVAE) is trained on multitemporal SAR composites to suppress speckle while preserving flood-relevant spatial structure. Experiments using UNet and UNet++ architectures demonstrate strong multispectral performance (AUPRC up to 0.956) and statistically significant improvements in SAR flood mapping when using shift-invariant loss and CVAE-based despeckling compared to classical filters. These results underscore the importance of dataset fidelity, misalignment-robust training, and demonstrate the viability of generative despeckling for operational flood mapping.
Read the full abstract

All authors David Ma, Jeremy Feinstein, Shreya Pandit, Arkaprabha Ganguli, Eugene Yan

Computer Vision

On the Faithfulness of Post-Hoc Concept Bottleneck Models

Laines Schmalwasser, Jan Blunk, Niklas Penzel et al. (+2 more)

Human decision-making interprets the world through high-level concepts, such as recognizing a bird by its belly color. To bridge the gap between opaque deep learning representations and human understanding, Post-Hoc Concept Bottleneck Models (post-hoc CBMs) project latent features onto interpretable concept spaces using auxiliary datasets or vision-language models. However, relying on target task accuracy as the primary measure of post-hoc CBM success obscures whether the learned concepts are semantically meaningful or merely predictive artifacts. For example, random concept projections can achieve competitive accuracy despite being semantically meaningless. In this work, we analyze the learned projections directly and identify two failure cases: First, for concept projections learned from auxiliary data, covariate shifts can lead to unfaithful concept representations for the target task. In particular, we provide an upper bound on the error introduced by this shift. Second, systematic label noise in surrogate concept labels generated by vision-language models leads to unfaithful projections. After formalizing these failure modes, we introduce novel metrics that decouple concept faithfulness from predictive accuracy. Our empirical results across real-world and synthetic benchmarks confirm that these metrics identify unfaithful behaviors that standard accuracy-based evaluation fails to detect.
Read the full abstract

All authors Laines Schmalwasser, Jan Blunk, Niklas Penzel, Julia Niebling, Joachim Denzler

Computer Vision

RBE-Flow: Recurrent Bayesian Estimation on Feature Manifolds for Cross-Modal Registration

Mengzhu Ding, Xin Song, Xiaoke Ding et al. (+2 more)

Cross-modal image registration is essential for multi-sensor perception but remains fundamentally challenging due to severe non-linear radiometric discrepancies and geometric distortions. Existing deterministic matching methods lack uncertainty awareness, struggling to navigate the resulting highly non-convex optimization landscape and frequently accumulating errors in ambiguous regions. In this paper, we propose RBE-Flow, a novel framework that reformulates dense cross-modal flow estimation as a closed-loop recurrent Bayesian estimation problem on learned feature manifolds. Diverging from standard feed-forward regression, RBE-Flow establishes a robust self-correcting mechanism by deeply coupling feature-metric non-linear optimization with probabilistic state updates. Specifically, a Recurrent Manifold Optimization (RMO) block iteratively generates flow observations and their associated uncertainties, which are then optimally assimilated into the prior state via an Uncertainty-Adaptive Probabilistic Update (UAPU) using deterministic sigma-point projection. Crucially, the resulting calibrated posterior covariance is fed back to adaptively regularize the damping of subsequent optimization steps, allowing the system to modulate its convergence based on predictive confidence. To ensure stable probabilistic training, we introduce a hybrid supervision scheme featuring a geometry-aware rectified NLL loss that structurally prevents variance collapse. Extensive experiments on challenging OSdataset, WHU-OPT-SAR, and RoadScene benchmarks demonstrate that RBE-Flow consistently achieves state-of-the-art performance, outperforming existing methods by a significant margin, particularly under strict sub-pixel criteria. Project page: https://github.com/NEU-Liuxuecong/RBE-Flow
Read the full abstract

All authors Mengzhu Ding, Xin Song, Xiaoke Ding, Hongwei Ding, Xuecong Liu

Computer Vision

PGE-SAM: Prompt-Guided Feature Enhancement for Interactive Segmentation under Degradation

Tuan-Duc Nguyen, Anh-Tuan Mai and Duc-Trong Le

Segment Anything Model (SAM) has revolutionized promptable image segmentation with strong zero-shot generalization. However, its performance degrades substantially under real-world imaging artifacts such as noise, blur, and compression. Existing methods restore features globally without focusing on segmentation-relevant regions and neglect SAM's iterative refinement mechanism, leading to suboptimal performance in interactive settings. We propose Prompt-Guided Feature Enhancement SAM (PGE-SAM), a framework that explicitly leverages user prompts and prior mask predictions to spatially guide the feature restoration process toward regions of interest through a Prompt Guidance Generator. To recover fine-grained details lost under degradation, we introduce Multi-Scale Features Interaction to incorporate low-level encoder features, along with a Foreground Reconstruction Loss that restricts feature-level supervision to the segmentation target. Furthermore, we present DM-Seg, a benchmark for interactive segmentation on degraded medical images, spanning multiple imaging modalities with both general and modality-specific degradations at varying severity levels. Extensive experiments demonstrate that PGE-SAM achieves SOTA robustness on both medical and natural image domains across multiple degradation levels, while maintaining generalization to clean images and adding less than one-fifth of the parameters of prior methods.
Read the full abstract

Computer Vision

PS-MOT: Cultivating Instance Awareness from Point Seeds for Multi-Object Tracking

Kai Luo, Fei Teng, Mengfei Duan et al. (+6 more)

We introduce Point-supervised Multi-Object Tracking (PS-MOT) as a cost-effective alternative to traditional bounding box supervision, shifting the focus from spatial fitting to topological center-driven representation. However, PS-MOT faces challenges, e.g., spatial ambiguity and identity drift due to the lack of explicit geometric structure and scale constraints. To address these, we propose PS-Track, a hierarchical pipeline transitioning from points to instances across data, model, and loss levels. At the data level, we introduce Temporal-Feedback Prompting (TFP) to evolve points into temporally consistent pseudo-labels using negative spatial cues and motion priors. At the model level, we design the Point-Excited Wavelet Attention (PEWA) module, which leverages semantic correlations to activate high-frequency components, “hallucinating” object boundaries. At the loss level, Uncertainty-Guided Gaussian Learning (UGL) models pseudo-labels as probabilistic distributions, dynamically calibrating supervision intensity. Experiments on DanceTrack, EmboTrack, SportsMOT, and JRDB demonstrate that PS-Track provides a feasible and effective point-supervised alternative across diverse tracking scenarios, establishing a new state-of-the-art for point-supervised tracking. The source code is available at https://github.com/xifen523/PS-MOT.
Read the full abstract

All authors Kai Luo, Fei Teng, Mengfei Duan, Wanjun Jia, Xu Wang, Hao Shi, Kunyu Peng, Zhiyong Li, Kailun Yang

Computer Vision

FR-DETR: Frequency and Recurrent Feature Refinement for Robust Object Detection under Adverse Weather

Tuan-Duc Nguyen and Duc-Trong Le

Object detection under adverse weather remains challenging due to severe visual degradations and domain shifts. Existing enhancer-based approaches attempt to improve detection by cascading an enhancer with a detector, but they introduce redundant feature extraction and incur high computational cost with limited accuracy gains when paired with SOTA detectors. We propose FR-DETR, a detector-centric framework that refines features rather than images, focusing enhancement on regions of interest and leveraging frequency-domain cues. Specifically, we design (I) a Frequency Refinement Module that dynamically separates and reweights low- and high-frequency components to improve foreground-background discrimination, and (II) a Recurrent Focus Refinement Module (RFRM) that iteratively refines features using coarse predictions as guidance. Extensive experiments demonstrate that FR-DETR achieves superior detection accuracy under adverse weather while being significantly more computationally efficient than enhancer-based methods. Our implementation is available at https://github.com/ducnt1210/FR-DETR.
Read the full abstract

Computer Vision

Cross-Resolution Semantic Transfer for Robust Text-to-Image Retrieval in Low-Resolution Surveillance

Wenjie Qian, Bin Yang, Xiao Wang et al. (+4 more)

Text-to-image person re-identification (TIPR) retrieves target persons using natural language descriptions. However, existing methods largely overlook resolution variance in real-world surveillance. They characterize cross-resolution TIPR through two coupled failure modes: Evidence Reliability Collapse (ERC), where degraded visual tokens become unreliable for grounding fine-grained text, and Ranking Distribution Drift (RDD), where mixed-resolution galleries distort similarity neighborhoods and destabilize retrieval rankings. To address this challenge, we propose Cross-Resolution Semantic Transfer (CRST), a CLIP-style framework with three modules: resolution-conditioned reasoning, text-guided refinement and CR-RDA. Resolution-conditioned reasoning estimates token reliability to suppress corrupted evidence. Text-guided refinement injects semantic priors to recover discriminative cues. CR-RDA transfers HR neighborhood geometry to stabilize LR ranking under mixed resolutions. Experiments on CUHK-PEDES, ICFG-PEDES, and RSTPReid show that CRST improves ultra-low-resolution Rank-1 and mAP on average by 5.7% and 5.3%, while stabilizing mixed-resolution retrieval without sacrificing high-resolution accuracy.The code will be made publicly available.
Read the full abstract

All authors Wenjie Qian, Bin Yang, Xiao Wang, Wenke Huang, Ling Mei, Xin Xu, Mang Ye

Computer Vision · Cross-listed from cs.RO

Vision-Language-Action Models: Experimental Insights from a Real-World UR5 Platform

Mathilde Hochedel and Marc Lalonde

This project investigates whether recent Vision-Language-Action (VLA) models can be transferred from controlled research benchmarks to a real-world robotic platform, specifically a UR5e manipulator, in a reproducible and operationally meaningful manner. The work integrates real-robot data acquisition, dataset engineering (compatible with the RLDS format), and the fine-tuning and deployment of OpenVLA and OpenVLA-OFT models, with systematic validation of action representations and control interfaces. The project resulted in several foundational assets: (i) a complete real-robot data acquisition pipeline, (ii) a dataset conversion workflow aligned with RLDS standards, (iii) an initial fine-tuning and inference infrastructure for VLA models, and (iv) a structured set of experimental observations grounded in real-robot trials. These elements collectively establish a reproducible framework for evaluating learning-based manipulation systems beyond simulation. Empirically, the experiments reveal a consistent gap between promising offline indicators and unstable closed-loop behavior on the physical system: this gap cannot be attributed solely to model limitations, it is strongly influenced by action semantics, coordinate frame conventions, temporal alignment between modalities, image preprocessing consistency, and dataset coverage and quality. These observations lead to a key interpretation: the successful deployment of VLA systems in real-world settings depends less on incremental improvements in model capacity and more on precise control of the entire data-model-control pipeline. The project reframes VLA-based robotics from a primarily model-centric challenge to a system-level problem; it highlights the difficulty of running robust task execution on the real robot and provides a clear, experimentally grounded understanding of the conditions required for reliable deployment.
Read the full abstract

Computer Vision

Robust and Efficient Monocular 3D Gaussian SLAM for Kilometer-Scale Outdoor Scenes

Sicheng Yu, Dongxu Shen, Beizhen Zhao et al. (+2 more)

Scaling monocular 3D Gaussian Splatting (3DGS) SLAM to kilometer-level outdoor environments poses two tightly coupled challenges: fragile long-term pose tracking and excessive memory overhead during large-scale mapping. In this paper, we propose KiloGS-SLAM, a highly efficient and robust monocular 3DGS-SLAM system that jointly addresses both bottlenecks. Since high-fidelity scene reconstruction fundamentally relies on drift-free camera poses, we first introduce a motion-adaptive hybrid tracking module. This module features a condition-triggered three-tier solving pipeline. It dynamically switches between Essential matrix and PnP models to handle geometric degeneracies. An on-demand foundation model can also be activated to rescue the trajectory from catastrophic drift. To ensure the system can sustain these long trajectories without memory exhaustion, we subsequently design a lifecycle-managed Gaussian mapping strategy. By integrating probabilistic initialization with chunk-based multi-view densification and pruning, this full-pipeline optimization effectively reduces primitive redundancy while preserving high-frequency details. Together, the robust tracking guarantees the geometric foundation required for accurate mapping, while the memory-efficient lifecycle-managed mapping enables large-scale operation. Extensive experiments across three challenging outdoor datasets demonstrate that our approach achieves state-of-the-art tracking accuracy and rendering quality, successfully scaling to sequences of over 10,000 frames on a single GPU.
Read the full abstract

All authors Sicheng Yu, Dongxu Shen, Beizhen Zhao, Guanzhi Ding, Hao Wang