SAUF-Net is a semi‑supervised medical image segmentation framework that learns structure–appearance representations with uncertainty feedback. It decomposes bottleneck features into structural and appearance components, injects them into decoding, and uses auxiliary decoders and a dual‑head discriminator to estimate reliability and uncertainty. Experiments on ISIC‑2016 and Kvasir‑SEG show that SAUF‑Net surpasses state‑of‑the‑art methods, particularly when few labels are available.
By Qin Lu, Zheyang Jing, Yujie Yang, Jianwang Li, Chen Yi, Shaofeng Jiang
InfraPatch is a white‑box, per‑instance framework that generates small grayscale patches to target infrared‑adapted vision‑language models (IR‑VLMs). The method optimizes a single‑channel patch within a 5% local‑area budget, using proxy‑guided placement and task‑adaptive objectives to induce desired behaviors in image classification, captioning, and binary visual question answering. Across ten IR‑VLM variants tested on synthetic infrared images, InfraPatch achieves targeted attack success rates ranging from 86% to 100%, revealing significant vulnerability differences among architectures and tasks.
By Chengyin Hu, Dingyi Lu, Jiaju Han, Xiang Chen, Weiwen Shi, Jiahuan Long, Yiwei Wei, Jiujiang Guo
InsightSeg introduces an episodic memory system that captures successful correction episodes from multi-agent refinement processes and transforms them into reusable, visually grounded insights. These insights are distilled into natural-language directives and linked to specific image patches via visual concept vectors, allowing the segmentation agent to anticipate and avoid guideline-specific errors on new images. The approach improves both first-pass and final segmentation quality on Waymo and Cityscapes datasets while reducing the number of required refinement steps.
By Vanshika Vats, Ashwani Rathee, James Davis
The paper presents a stereo 4D Radar framework for 3D object detection that uses geometric disparity between left and right radars to estimate absolute velocity and fuse complementary features. It addresses clutter, ghost reflections, and sparse data issues inherent in raw 4D Radar signals. Experiments on an in‑house dataset show significant gains, improving AP 3D by 8.82 points and AP BEV by 9.0 points over mono‑radar baselines.
By Seung-Hyun Song, Dong-Hee Paek, Woong-Chan Byun, Seung-Hyun Kong
T2LSC-Bench is a new benchmark for evaluating localized semantic control in text-to-image generation, consisting of 50 seed subjects and 1,200 prompt cases per model, producing 7,160 images across six models. The benchmark measures Text-at-Anchor Accuracy, Semantic Subject Preservation, Semantic Leakage Rate, and Conditional Semantic Leakage Rate using a dual‑branch protocol that combines OCR‑VLM verification with structured VLM semantic judgments. Results show that while accurate text rendering remains high, semantic leakage can increase dramatically under stress‑test conditions, and anti‑leakage prompting can reduce leakage without harming rendering accuracy.
By Yan Wang, Xinyi Hou, Weiguo Lin, Junjun Si, Siwei Ma
The paper introduces a two-stage framework for point-supervised change detection that leverages SAM2 priors to generate object-aware candidate masks and refines them with a lightweight CNN and uncertainty-aware loss. In the second stage, a teacher‑student self‑training loop with exponential moving average updates continuously improves pseudo‑labels and model performance. Experiments on WHU-CD, LEVIR-CD, and SYSU-CD show the method surpasses prior weakly supervised approaches and competes with fully supervised ones.
By Hailong Ning, Hao Wang, Yimeng Wang, Tao Lei, Renwei Dian, Asoke K. Nandi
DESA‑TTA introduces a low‑overhead test‑time adaptation technique for vision‑language object detectors that dynamically adjusts the teacher’s exponential moving average coefficient based on pseudo‑label confidence and box density. It also employs source anchoring to gradually pull student parameters back toward their pretrained values, mitigating cumulative drift. Experiments on multiple distribution shifts and two VLOD architectures demonstrate consistent gains, with a 14.5‑point AP₅₀ improvement on VOC‑C and a 55% higher inference throughput compared to the prior state‑of‑the‑art TTA method for YOLO‑World.
By Atif Belal, Lilian Hollard, Marco Pedersoli, Eric Granger
CoViT introduces a self‑supervised framework that enhances Vision Transformers with instance‑aware representations by leveraging geometry‑guided contrastive learning. It refines attention maps to generate instance masks and constructs triplets that mine the hardest intra‑ and inter‑instance examples, driving a contrastive loss that reduces intra‑instance variance while increasing inter‑instance margins. The method yields consistent AP gains of over 2 points on instance‑level tasks without requiring extra decoders or labels.
By Yisen Wang, Zhirong Wu, Limin Wang
LeakageBench is a new benchmark comprising 500 document images with 11,954 GDPR‑aligned PII annotations, designed to evaluate document‑level redaction risk. It measures how well OCR pipelines, OCR‑dependent detectors, and OCR‑free vision‑language models can localize and remove sensitive information, using entity‑level F1, group‑wise leakage, and document‑level leakage metrics. The study shows that while advanced models improve localization, most pages still exhibit critical leakage, highlighting the need for higher‑recall, spatially grounded redaction methods.
By Vishnu Prasad Vijaya Kumar, Santhosh Venkatesh, Ivan P. Yamshchikov
The paper presents a hardware‑accelerated instance segmentation framework tailored for resource‑constrained lunar robotics, addressing low‑light perception, limited compute, and radiation‑induced hardware faults. It introduces Activation Variance Informative Sampling (AVIS), a label‑free calibration method that selects samples based on activation variance, and deploys a YOLO‑based model on a Deep Learning Processor Unit with architectural tweaks to reduce CPU fallback and ensure bounded latency. A software‑level criticality analysis estimates fault exposure, guiding mitigation that reduces global criticality by 31.7%, while AVIS with bias correction recovers 69.8% of quantization‑induced accuracy loss at 309 ms latency and 5.7 W power consumption.
By Siddhant Shete, Hilmi Dogu K\"uc\"uker, Udo Frese, Frank Kirchner
The paper critiques the common practice of random pixel splits in hyperspectral image classification, noting that such splits allow test pixels to be adjacent to training pixels, inflating accuracy. It proposes a leakage‑free evaluation protocol that enforces spatial separation based on the model’s receptive field and applies it to ten diverse architectures, finding a significant drop in Macro‑F1 (average 0.147) and substantial changes in model rankings. The study also shows that all ten models misclassify the same pixels, indicating a spectral ambiguity in the data that current methods cannot resolve.
By Ehsan Faghih, Fatemeh Ashrafi, Marguerite Moore, Zahra Saki
The paper investigates the teacher‑student framework used in Test‑Time Adaptation (TTA) and questions the common practice of updating the teacher via an exponential moving average of the student. The authors demonstrate that error accumulation still occurs, especially over longer sequences, and propose an intransigent teacher that remains fixed. This modification yields significant performance gains across multiple datasets, longer scenarios, and various architectures, including semantic segmentation, while also improving robustness to hyperparameter changes.
By Damian S\'ojka, Marc Masana, Bart{\l}omiej Twardowski, Sebastian Cygert
VTOS (Vision Tools Orchestration Search) is a framework that adaptively orchestrates vision foundation tools—such as open‑vocabulary detectors, segmentation models, and post‑processing operators—by jointly searching for executable solution programs and observer programs that diagnose failures and provide feedback. The observer programs feed observations into a shared VisionThoughts knowledge base, guiding subsequent searches. In two case studies—dense object counting on LVIS‑Count and zero‑shot plant‑disease segmentation on PlantSeg‑OOD—VTOS outperforms static tool pipelines and agentic visual‑programming baselines, especially in complex scenarios like dense, occluded scenes and out‑of‑distribution segmentation.
By Jinchao Ge, Lingqiao Liu, Shuwen Zhao, Lei Wang
VeriPhy is an auditable physical‑verification system that evaluates generated video by compiling prompts into typed physical obligations and a statically validated execution plan before any frames are observed. During execution, it gates calls to frozen low‑level experts (e.g., segmentation, tracking, counting, depth, OCR, audio‑event detection) and returns provenance‑carrying evidence records, which are mapped to a three‑valued state (supported, contradicted, unknown) with full traceability. On a 1,500‑clip corpus of human‑annotated flaw records, VeriPhy accounts for 228 failures out of 304, outperforming a published question‑decomposition evaluator that accounts for 164, while also providing auditable evidence for each verdict.
The paper presents a stereo 4D Radar-based framework for 3D object detection that uses the geometric disparity between left and right radars to estimate absolute velocity and fuse complementary features. It addresses challenges such as clutter, ghost reflections, and sparse data caused by preprocessing, and improves motion state estimation beyond the radial Doppler component. Experiments on an in‑house stereo 4D Radar dataset show significant gains of 8.82 points in AP 3D and 9.0 points in AP BEV over mono‑radar baselines.
The paper introduces Doppio, a new video dataset of falling ground coffee with precise per‑frame weight measurements, aimed at enabling contactless mass estimation. It evaluates several deep learning models—spatial feed‑forward and recurrent spatio‑temporal networks—to predict the cumulative weight of falling particles. The results show that these vision‑based approaches can accurately estimate weight, suggesting a practical foundation for future contactless measurement systems.
WiFlow introduces a method for estimating optical flow using WiFi channel state information (CSI) instead of camera frames, addressing privacy and lighting issues. The approach includes a CSI-based flow estimator, a preprocessor evaluation, and three model architectures balancing accuracy and complexity. Additionally, the authors provide the first dataset for training and evaluating CSI-based optical flow estimators, along with experimental insights into key design elements.
The paper introduces the concept of information density to explain category bias in visual object detection. It finds a strong negative correlation between a category’s information density and its detection accuracy, showing that instance count alone does not account for bias. By incorporating information density into three advanced loss functions, the authors demonstrate significant bias reduction and overall performance gains on Pascal VOC, COCO‑LT, and LVIS datasets.
LeakageBench is a new benchmark consisting of 500 document images with 11,954 GDPR‑aligned PII annotations, designed to evaluate document‑level redaction risk. Unlike existing text‑centric PII benchmarks, it measures whether a page remains unsafe if any identifier is missed, using entity‑level F1, group‑wise leakage, and document‑level leakage metrics. Experiments show that even advanced OCR pipelines and vision‑language models improve localization but still leave a high proportion of pages unsafe for release.
The paper introduces a two‑stage framework for point‑supervised change detection that leverages SAM2 priors to generate object‑aware candidate masks from sparse point annotations. In Stage I, a mask selection strategy converts generic segmentation outputs into reliable change pseudo‑labels, followed by a lightweight CNN refinement module with an uncertainty‑aware loss to enhance boundary quality. Stage II employs a teacher‑student self‑training loop, where the teacher is updated via exponential moving average and periodically refreshes pseudo‑labels, creating a closed‑loop optimization that alternates between pseudo‑label refinement and model re‑optimization. Experiments on WHU‑CD, LEVIR‑CD, and SYSU‑CD show the method surpasses prior weakly supervised approaches and competes with several fully supervised methods.