Searcharxiv⌕ Search

arXiv subjects

Computer vision

Explore arXiv computer-vision papers and cs.CV metadata. Search within the index for image segmentation, object detection and visual representation learning.

Follow this topic with RSS
At least 1,243 records · Page 69Linked to original sources
arXiv2026

MeshSplatBench: A Unified Benchmark for Triangle- and Mesh-Based Neural Rendering

Triangle- and mesh-based neural rendering aims to bridge neural scene representations and existing graphics engines (\textit{e.g.}, Unity and Blender) by leveraging triangle primitives compatible with standard rasterization hardware. However, existing methods are developed and evaluated under inconsistent settings, with limited comparison and little investigation into practical graphics engine deployment. This gap significantly hinders the understanding of their real-world usability. To address this issue, we introduce MeshSplatBench, the first benchmark for systematic evaluation of triangle- and mesh-based neural rendering from native rendering to graphics engine deployment. We propose a hierarchical deployment protocol with two options: (1) Standard deployment, using a conventional opaque mesh pipeline with vertex colors and hardware Z-buffering; and (2) Dedicated deployment, incorporating method-specific engine implementations to preserve appearance and compositing properties (e.g., alpha blending). For mesh splatting, we further introduce a structural audit to evaluate the topological and geometric integrity of exported surfaces for downstream graphics applications. Extensive evaluations reveal three key findings: (1) graphics engine deployment introduces noticeable quality degradation across methods, while mesh splatting approaches achieve relatively better robustness under standard deployment; (2) dedicated deployment can preserve most rendering fidelity at the cost of approximately 6-30$\times$ slowdown; and (3) explicit connectivity and shared vertex indexing in current mesh splatting methods remain insufficient to guarantee manifoldness or global connectivity. Our benchmark demonstrates that rasterizability alone does not imply graphics readiness and highlights the importance of evaluating practical engine compatibility. The benchmark and source code will be publicly released.

cs.GR↗
arXiv2026

SimFuse3D: Source-Guided Target Simulation and Confidence-Guided Multi-Stage Localization Reweighting for Cross-Platform 3D Object Detection

Changes in sensor height and viewpoint alter object-level point distributions, making cross-platform LiDAR unsupervised domain adaptation (UDA) difficult. Self-training uses labeled source scans and unlabeled target scans, yet a retained prediction may provide a useful target location while enclosing sparse foreground returns, background clutter, or points inconsistent with the predicted box. We refer to this mismatch as box-point inconsistency. We introduce SimFuse3D, which preserves the target placement and repairs the associated pseudo object using measured geometry from labeled source scans. Object Memory retrieves a similar labeled source instance. Target Simulation places the retrieved source geometry at the target location, aligns its points with the target viewing geometry, and filters the aligned crop to approximate the target observation. Confidence-Guided Multi-Stage Localization Reweighting (CMLR) maps each target pseudo-object confidence score to a bounded weight shared by RPN localization and R-CNN box regression. All components operate only during adaptation, leaving the detector architecture and inference graph unchanged. Across six cross-platform transfers, SimFuse3D consistently outperforms Pi3DET-Net and achieves the best performance among the compared adaptation methods on nearly all metrics. On nuScenes-to-KITTI, it ranks first among the compared adaptation methods with both evaluated detectors.

cs.CV↗
arXiv2026

AcFlow: Controlling Text-to-Image Diffusion Transformers via Learned Conditional Activation Flow

Text-to-image diffusion transformers (DiTs) are powerful generators, yet direct prompting provides limited control interface for style intensity and can fail to suppress unwanted concepts. To enable these controls, we introduce AcFlow, an inference-time controller that transports intermediate layer image-token activations through a learned concept-conditioned velocity field while keeping the base DiT frozen. A textual concept description specifies the desired intervention, while the integration horizon provides a continuous control parameter. The field produces token-varying, activation-dependent updates. With parameters shared across concepts within each task family, one field covers over 15,000 style descriptions or over 1,000 suppression concepts, and generalizes to concepts unseen during training without per-concept fitting. On style control, AcFlow achieves the best style--content trade-off among the evaluated baselines in the high-style-alignment regime. At a fixed operating point, AcFlow attains style--content alignment of 0.5365/0.2860, compared with 0.4397/0.2684 for the baseline with the highest style alignment. On concept suppression, AcFlow reduces the fraction of images showing the concept from 95.3%/82.1% to 41.6%/40.5% on held-in/held-out concepts, including cases where deleting them from the prompt fails to remove them. Our analyses support the learned velocity field as an adaptive control mechanism, with update directions varying across tokens and depending on their activation states. Our code is available at https://github.com/Nove1yst/AcFlow.

cs.CV↗
arXiv2026

Decodable but Misrouted: Sparse Features Uncover a Readout Gap in Vision-Language Models for Harmful Meme Detection

When large vision-language models misclassify harmful memes, the failure may reflect missing internal evidence or an inability to route represented evidence to their outputs. We distinguish these cases in Gemma-3 and Qwen3.5 using sparse autoencoders, role-conditioned probes, causal interventions, and recovery experiments across six harmful content benchmarks, with additional Spanish and Hindi-English code-mixed evaluations. Sparse readouts outperform native prediction on all six primary binary tasks: Qwen averages $0.740$ versus $0.432$ for native macro-F1, residual reconstruction reaches $0.486$, and Gemma improves from $0.532$ to $0.714$. These gains measure how accessible the label is to a supervised readout; they do not show that the model's native generation already applies such a decision rule. Under the evaluated scales, Qwen silent-feature ablation is $24-63$ times more probe-sensitive, whereas routed-feature patching on literal yes/no tasks is $16-140$ times more output-sensitive. Native-only threshold calibration explains much, but not all of the gap: on five tasks with matched probe scores, it recovers $69.8$\% of the raw native-to-probe difference, while direct routing adds $0.094$ mean macro-F1 beyond calibrated native scoring. Joint gold-label, probe-KL, and pairwise LoRA supervision improves dedicated FHM prediction, but a gold-only adapter performs better on the shared seven-task mean. A case study of Gemma-3-12B on the Facebook Hateful Memes dataset finds a distributed rank-32 image-prompt interaction, reaching $0.756$ versus $0.685$ native macro-F1. Robustness controls show that the signal is not explained solely by accompanying OCR and depends on paired visual evidence, and that it extends beyond English. In many of the errors we study, the evidence is represented but does not reach the answer; therefore, routing is a common bottleneck in harmful meme classification.

cs.CV↗
arXiv2026

RoXDrive: Closed-Loop Reinforcement Learning for End-to-End Autonomous Driving via Action-Faithful Rollouts

End-to-end autonomous driving policies are commonly trained via imitation learning on logged demonstrations without observing the consequences of their own actions, leading to causal confusion in closed-loop real-world deployment. To address this issue, reinforcement learning (RL) post-training offers a promising alternative by leveraging world models as interactive training environments to enable future scene generation for policy improvement. Nevertheless, existing approaches either rely on reconstruction-based simulators, offering limited counterfactual interaction, or adopt synthetic simulators to enable long-horizon closed-loop interaction at the cost of a substantial sim-to-real gap. Recently, video world models have exhibited the ability to generate realistic multi-step future rollouts but may not faithfully reflect action conditions, resulting in action-vision mismatch. In this paper, we introduce RoXDrive, a plug-and-play closed-loop RL framework that enables reliable policy optimization by identifying action-faithful world-model rollouts, consisting of two stages: 1) Model pre-training: In addition to imitation-based policy pre-training, we devise an Action-Vision Faithfulness Evaluator for inverse dynamics estimation with our geometry-aware auxiliary trajectory supervision, enabling long-horizon assessment of whether visual dynamics faithfully reflect the conditioning ego actions. 2) Action-faithful RL post-training: Agents iteratively interact with world models to form long-horizon scene rollouts, retaining only action-faithful ones for dense safety-aware scoring and scene-level closed-loop RL post-training. Extensive experiments on nuScenes and an in-house dataset with over 130K training scenarios demonstrate consistent gains across planners, reducing safety violations by 27.6% with DiffusionDrive on nuScenes and 33.7% with Qwen3-VL on the internal data.

cs.CV↗
arXiv2026

PAIQ: Patch-Aligned Semantic Injection via Residual Rotation

Language-aligned and self-supervised visual encoders offer complementary strengths in semantic abstraction and spatial detail. Harnessing this complementarity requires enriching local features while retaining distinctions between semantically related patches. We introduce PAIQ, a patch-aligned semantic injection framework that combines content-based cross-encoder matching with orthogonally constrained residual updates. Using DINOv3 patch features as the spatial base, PAIQ aggregates complementary SigLIP features through joint source allocation and injects the aggregate--base differences through a shared orthogonal transformation Q. This rotation adapts update directions while preserving residual norms and pairwise angles. For fixed projected features, we derive conditions for patch separability under similar semantic aggregates and show that rotation adds a nonnegative separation term over direct interpolation when the aggregate is shared. Only the projection and fusion parameters are trained; both visual encoders and the language model remain frozen, and fusion retains 196 visual tokens. Across diverse language backbones, PAIQ yields broad gains in judge-assessed correctness and reductions in hallucination severity over single-encoder interfaces on image description and visual question answering. On the 2B and 9B Qwen backbones, this compact interface outperforms the strongest evaluated fusion or token-compression baselines by about 2.9 correctness points on average.

cs.CV↗
arXiv2026

Honeycomb: Constant-Size Scene Memory Representation for Video World Models

Video world models require persistent scene memory to maintain consistency during long-horizon video generation. Existing spatial memory systems accumulate RGB observations or latent features, causing storage requirements to grow as generation proceeds. We introduce **Honeycomb**, a video world model built on **HexMemory**, a compact low-rank representation that stores scene features in a fixed-size memory comprising six spatial and spatiotemporal planes. A feed-forward writer maps each newly generated video chunk to plane features. As the spatial coverage or temporal range expands, HexMemory warps the existing planes while preserving their dimensions, then integrates new features through confidence-weighted pooling and a learned residual correction. A reader retrieves latent features from HexMemory to condition subsequent video generation. Because the writer processes only observations from the latest chunk, Honeycomb avoids per-scene optimization and repeated processing of the full generation history. Experiments on WorldScore and RealEstate10K demonstrate strong video generation quality and robust consistency when revisiting previously observed regions, while maintaining constant feature-storage requirements throughout generation. Code and additional visualizations are available on our https://jackswl.github.io/honeycomb/.

cs.CV↗
arXiv2026

VIF-Bench: Evaluating Visual Instruction Following in Multi-Reference Image Generation

Recent multimodal image generation models can take multiple images and textual instructions as input, enabling reference-based generation guided not only by text but also by visual instructions such as layouts, arrows, and pose cues. However, existing benchmarks do not evaluate the joint setting in which multiple references must be composed under multiple and heterogeneous visual-instruction images. To address this gap, we introduce VIF-Bench, a benchmark of 1,241 tasks designed to assess the edge of model capabilities in this joint setting by covering: (i) multi-reference generation (up to 7) under multiple heterogeneous visual instructions (up to 6), (ii) cases where reference images can potentially compete with visual instructions (e.g., a strongly posed subject vs. a target pose), and (iii) controlled comparison of visual instructions with text descriptions at different levels of specificity. Using these capabilities, we uncover three findings: (1) models face an adherence-artifact trade-off: once models reach stronger visual instruction adherence, stronger adherence tends to coincide with more instruction artifacts in generated images, (2) visual instruction adherence tends to be lower on tasks whose reference images carry a salient state of the controlled attribute (e.g., a neon-lit subject under a light-direction instruction), most consistently for light and wind, and (3) for models that can understand visual instructions, it is often better to provide visual constraints directly rather than describe them in text; when using text, a moderate level of detail works better than an exhaustive description. VIF-Bench is released as an open benchmark to establish a basis for fair comparison in controllable multi-reference image generation.

cs.CV↗
arXiv2026

PolyOCR-Venus: Unified OCR Foundation Models for Text-Centric Visual Intelligence

Optical Character Recognition (OCR) is evolving from plain-text transcription toward general visual intelligence, requiring models to recognize, localize, and reason over textual information in complex visual environments. However, existing OCR systems often excel at only some tasks and struggle to balance recognition, parsing, and reasoning across scenarios. In this report, we present PolyOCR, a family of unified OCR foundation models of varying scales. PolyOCR combines a shared instruction-following framework with a large-scale data engine that converts heterogeneous visual resources into quality-verified OCR supervision. We introduce Competence-Guided Policy Optimization, which combines verifier-based Group Relative Policy Optimization with on-policy distillation through sample-wise routing based on teacher reliability and the teacher--student competence gap. We also introduce OCRBench v2.1, our revision of OCRBench v2 with manually verified annotation corrections and task-aligned scoring metrics. Extensive experiments across OCRBench v2.1, CC-OCR, in-house KIE Benchmark, OmniDocBench v1.6 and MDPBench demonstrate that PolyOCR achieves state-of-the-art or highly competitive performance.

cs.CV↗
arXiv2026

CogWAM: Aligning Semantic Cognition with World Action Modeling via Event-Driven Interfaces

Robot policies increasingly incorporate semantic reasoning and future-world prediction, yet combining these capabilities does not guarantee that local predictions and actions remain aligned with task progress. We introduce CogWAM, a cognition-guided world-action model that establishes an explicit semantic interface between task reasoning and world-action learning through a persistent Semantic State, which stores completed task events and the active subtask. CogWAM updates this state only when observations indicate semantic transitions, allowing task-level context to persist across multiple action chunks. To bridge semantic context with physical prediction and control, CogWAM employs progress-conditioned WORLD and ACTION queries that selectively extract task-relevant information for future-world prediction and action generation. During training, the Semantic State provides shared task-progress context for both branches, while inference removes the future-prediction branch and directly generates actions from observations and the maintained state. We further introduce semantic training strategies to improve transition learning and closed-loop conditioning. Without additional robot-action pretraining, CogWAM achieves 15.56 / 11.70 % Score/SR on RoboDojo and state-of-the-art performance on BiCoord, while real-world experiments demonstrate closed-loop dual-arm manipulation with 16.4 fewer Semantic State regenerations than step-wise updating.

cs.RO↗
arXiv2026

The Camera Inside the Editor: Reading the Implicit Camera of Image Editors with Painted Calibration Patterns

Instruction-based image editors insert objects, restyle scenes and render new viewpoints, but it is unknown which camera they assume when they paint into a photograph. Asked to cover the floor with a checkerboard, an editor paints projective structure from which classical vanishing-point geometry reads pitch, roll, focal length, yaw and, on renders, the principal point, without any training. Unlike a calibrator such as GeoCalib, which estimates the camera of an image, this isolates the camera under which the editor paints. On 120 rendered cameras with exact ground truth, Qwen-Image-Edit-2511 paints tile edges that meet their vanishing points within 0.26 degrees, and its implicit camera matches the true one to 0.8 degrees in pitch and 6% in focal length, more accurately than GeoCalib except in roll. Asked to draw the horizon or mark a vanishing point instead, the editor fails, so this knowledge is revealed by painting and not by the explicit tasks we tried. The implicit camera has two priors: roll is pulled towards level (slope 0.71), and telephoto perspective towards a default of about 30 mm, which roughly matches the camera the models paint without any scene. For Qwen, the priors do not grow when blur removes four fifths of the line evidence. They are stronger on real photographs, and on NYUv2 a shorter wording of the task removes the difference for roll. On photographs from a 24--240 mm zoom lens the painted perspective grows with only 0.62 of the lens's slope, while GeoCalib and MoGe-2 saturate at about 52 and 42 mm. FLUX.1 Kontext and LongCat-Image-Edit are pulled much harder. Finally, from a level camera a camera-control LoRA executes pose commands at only 50--70% of their strength, and a board painted into its output agrees with the camera it produced.

cs.CV↗
arXiv2013

Contextually learnt detection of unusual motion-based behaviour in crowded public spaces

In this paper we are interested in analyzing behaviour in crowded public places at the level of holistic motion. Our aim is to learn, without user input, strong scene priors or labelled data, the scope of "normal behaviour" for a particular scene and thus alert to novelty in unseen footage. The first contribution is a low-level motion model based on what we term tracklet primitives, which are scene-specific elementary motions. We propose a clustering-based algorithm for tracklet estimation from local approximations to tracks of appearance features. This is followed by two methods for motion novelty inference from tracklet primitives: (a) we describe an approach based on a non-hierarchial ensemble of Markov chains as a means of capturing behavioural characteristics at different scales, and (b) a more flexible alternative which exhibits a higher generalizing power by accounting for constraints introduced by intentionality and goal-oriented planning of human motion in a particular scene. Evaluated on a 2h long video of a busy city marketplace, both algorithms are shown to be successful at inferring unusual behaviour, the latter model achieving better performance for novelties at a larger spatial scale.

cs.CV↗
arXiv2025

BreastSegNet: Multi-label Segmentation of Breast MRI

Breast MRI provides high-resolution imaging critical for breast cancer screening and preoperative staging. However, existing segmentation methods for breast MRI remain limited in scope, often focusing on only a few anatomical structures, such as fibroglandular tissue or tumors, and do not cover the full range of tissues seen in scans. This narrows their utility for quantitative analysis. In this study, we present BreastSegNet, a multi-label segmentation algorithm for breast MRI that covers nine anatomical labels: fibroglandular tissue (FGT), vessel, muscle, bone, lesion, lymph node, heart, liver, and implant. We manually annotated a large set of 1123 MRI slices capturing these structures with detailed review and correction from an expert radiologist. Additionally, we benchmark nine segmentation models, including U-Net, SwinUNet, UNet++, SAM, MedSAM, and nnU-Net with multiple ResNet-based encoders. Among them, nnU-Net ResEncM achieves the highest average Dice scores of 0.694 across all labels. It performs especially well on heart, liver, muscle, FGT, and bone, with Dice scores exceeding 0.73, and approaching 0.90 for heart and liver. All model code and weights are publicly available, and we plan to release the data at a later date.

eess.IV↗
arXiv2026

HiReFF: High-Resolution Feedforward Human Reconstruction from Uncalibrated Sparse-View Video

Uncalibrated volumetric video streaming for human reconstruction is essential for holographic communication and AR/VR, yet remains challenging due to the need for temporal consistency and computational efficiency from sparse-view inputs. Existing methods rely on per-scene optimization or calibrated cameras, while recent feed-forward models are limited to low-resolution (0.5K) single-frame synthesis. We present HiReFF, a feed-forward method for 2K-resolution 360° human video reconstruction from uncalibrated sparse-view videos. Our framework decomposes the problem into two key tasks: foreground 3D Gaussian reconstruction from sparse-view videos (four views separated by 90°) and computationally efficient high-resolution synthesis. To enable the former, we propose Scale-synchronized Camera Calibration to resolve scale ambiguity for multi-view supervision, and Gaussian-wise Foreground Masking to reconstruct clean foregrounds by modulating Gaussian parameters. For efficient high-resolution synthesis, our High-resolution Side-tuning achieves 2K rendering by augmenting the Gaussian head with supplementary features while keeping the backbone at 0.5K, drastically reducing computational overhead. Experiments demonstrate that HiReFF significantly outperforms existing methods in high-resolution streaming volumetric video reconstruction. https://iridescentjiang.github.io/HiReFF

cs.CV↗
arXiv2026

DeepForestVisionV2: Ecology-Driven Taxonomy Expansion for Camera-Trap Monitoring in African Tropical Forests

Camera-trap monitoring in African tropical forests increasingly extends beyond closed-canopy interiors to riverbanks, clearings, and park edges. Among available open tools for African forest camera-trap classification, DeepForestVision is the only one providing a matched offline workflow for both photographs and videos, and previous work showed that it outperformed other available baselines on a comparable benchmark. However, it was designed for closed-canopy, ground-level forest interiors and uses a 35-class prediction space that becomes too coarse when deployments encounter arboreal primates, birds, semi-aquatic taxa, or human-associated confounders such as livestock. We present DeepForestVisionV2, an ecology-driven expansion from 35 to 64 prediction classes (61 animal classes plus human, vehicle, and blank) designed to address three recurrent deployment gradients: vertical stratification, scene openness, and anthropogenic interfaces. DeepForestVisionV2 retains the same offline workflow and is trained on 1,535,010 photographs and 243,354 videos from multi-country African tropical-forest projects. Evaluation combines a cross-country cropped-photo validation set, used to assess robustness across sites and camera-trap settings, with three held-out Uganda video benchmarks spanning the targeted gradients. On the validation set, DeepForestVisionV2 reaches 0.86 accuracy, 0.82 macro-F1, and 0.81 balanced accuracy. On the deployment benchmarks, it preserves or improves baseline accuracy despite its harder classification task, while increasing the number of identified taxa from 22 to 29 in forest-interior videos and from 4 to 9 at riverbanks. In the park-edge use case, it raises accuracy from 0.62 to 0.86 and reduces false alarms from 11 to 0. These results show that DeepForestVisionV2 materially improves field utility while preserving robustness across sites, habitats, and camera-trap settings.

cs.CV↗
arXiv2013

Multiple-object tracking in cluttered and crowded public spaces

This paper addresses the problem of tracking moving objects of variable appearance in challenging scenes rich with features and texture. Reliable tracking is of pivotal importance in surveillance applications. It is made particularly difficult by the nature of objects encountered in such scenes: these too change in appearance and scale, and are often articulated (e.g. humans). We propose a method which uses fast motion detection and segmentation as a constraint for both building appearance models and their robust propagation (matching) in time. The appearance model is based on sets of local appearances automatically clustered using spatio-kinetic similarity, and is updated with each new appearance seen. This integration of all seen appearances of a tracked object makes it extremely resilient to errors caused by occlusion and the lack of permanence of due to low data quality, appearance change or background clutter. These theoretical strengths of our algorithm are empirically demonstrated on two hour long video footage of a busy city marketplace.

cs.CV↗
arXiv2025

A Novel Combined Optical Flow Approach for Comprehensive Micro-Expression Recognition

Facial micro-expressions are brief, involuntary facial movements that reveal hidden emotions. Most Micro-Expression Recognition (MER) methods that rely on optical flow typically focus on the onset-to-apex phase, neglecting the apex-to-offset phase, which holds key temporal dynamics. This study introduces a Combined Optical Flow (COF), integrating both phases to enhance feature representation. COF provides a more comprehensive motion analysis, improving MER performance. Experimental results on CASMEII and SAMM datasets show that COF outperforms single optical flow-based methods, demonstrating its effectiveness in capturing micro-expression dynamics.

cs.CV↗
arXiv2025

GaussianFocus: Constrained Attention Focus for 3D Gaussian Splatting

Recent developments in 3D reconstruction and neural rendering have significantly propelled the capabilities of photo-realistic 3D scene rendering across various academic and industrial fields. The 3D Gaussian Splatting technique, alongside its derivatives, integrates the advantages of primitive-based and volumetric representations to deliver top-tier rendering quality and efficiency. Despite these advancements, the method tends to generate excessive redundant noisy Gaussians overfitted to every training view, which degrades the rendering quality. Additionally, while 3D Gaussian Splatting excels in small-scale and object-centric scenes, its application to larger scenes is hindered by constraints such as limited video memory, excessive optimization duration, and variable appearance across views. To address these challenges, we introduce GaussianFocus, an innovative approach that incorporates a patch attention algorithm to refine rendering quality and implements a Gaussian constraints strategy to minimize redundancy. Moreover, we propose a subdivision reconstruction strategy for large-scale scenes, dividing them into smaller, manageable blocks for individual training. Our results indicate that GaussianFocus significantly reduces unnecessary Gaussians and enhances rendering quality, surpassing existing State-of-The-Art (SoTA) methods. Furthermore, we demonstrate the capability of our approach to effectively manage and render large scenes, such as urban environments, whilst maintaining high fidelity in the visual output.

cs.CV↗