Detection, segmentation, depth and recognition research, plus the vision backbones that keep displacing the last generation.
arXiv:2608.15698v2 Announce Type: replace Abstract: Visual document retrieval is a critical component of multimodal retrieval-augmented generation, aiming to identify query-relevant pages from docume...
arXiv:2608.21133v1 Announce Type: new Abstract: Medical image-text data can expose protected health information (PHI) through both visible image content as well as accompanying text, creating a barri...
arXiv:2608.20942v1 Announce Type: new Abstract: Motivated by the classical Chan-Vese model and the ability of deep priors to capture complex spatial structures, we develop a segmentation model that l...
arXiv:2608.11645v2 Announce Type: replace Abstract: Volumetric video streaming turns privacy into a 3D, multi-view problem. Unlike ordinary video, where sensitive content can often be redacted frame...
arXiv:2608.21099v1 Announce Type: cross Abstract: Multi-modal object detection is essential for robust scene understanding in challenging conditions, including low-light and adverse environments. Rec...
The paper surveys Continual Test-Time Adaptation (CTTA), a framework that adapts pretrained computer‑vision models to non‑stationary target distributions without source data or labeled targets, while avoiding catastrophic forgetting and error accumulation. It formally defines the CTTA problem, categorizes existing methods into optimization‑based, parameter‑efficient, and architecture‑based families, and reviews representative techniques and benchmarks across standard evaluation settings. The survey also outlines current limitations and proposes future research directions, such as adapting foundation models and black‑box systems.
The paper presents the first real‑world 6D pose ground‑truth dataset for red‑stage strawberries, collected from 12,040 images at an actual farm using indirect camera pose recovery and 3D bounding‑box annotation. It also introduces a synthetic dataset rendered in NVIDIA Isaac Sim with scene‑level realism and domain randomization. Experiments show that models trained solely on synthetic data do not transfer well to in‑field images, but adding a small amount of real data significantly improves both translation and rotation accuracy across various backbone encoders.
This study introduces a two‑stage vision‑language model framework to assess the clinical quality and usability of late gadolinium enhancement (LGE) cardiac MRI images used for atrial fibrillation ablation planning. The first stage employs a fine‑tuned VLM to generate structured radiology‑style reports on five quality criteria—Noise, Motion Artifact, LA Boundary Accuracy, PV Region Accuracy, and Under‑segmentation Severity—while the second stage uses a GPT‑based reasoning module to convert these reports into structured quality scores and a binary decision on clinical usability. Evaluated on a curated dataset of 60 image‑slice and text‑pair annotations from 20 patients, the InternVL2 model achieved the highest criterion‑level accuracy, and DeepSeek reached perfect agreement on the clinical usability decision.
The paper introduces Semantic Localization-Enhanced Teacher (SLE‑T), a knowledge‑distillation framework that aligns spatial‑scale and semantic features between a Vision Foundation Model (VFM) teacher and a student detector for cross‑domain object detection. SLE‑T employs a lightweight SLE Adapter that injects pretrained local‑texture priors into DINOv2 and reformulates its features into dense, spatially and semantically compatible representations, enabling effective pseudo‑label learning or feature alignment. Experiments on three domain‑adaptive object detection benchmarks show that SLE‑T with DINOv2‑B achieves state‑of‑the‑art performance while using only a quarter of the training time and less GPU memory compared to the larger DINOv2‑G teacher.
The paper introduces a multi‑modal traffic sign detection framework that fuses camera and LiDAR data using an Intensity‑Aware Deformable Fusion module to align retro‑reflective LiDAR cues with visual features. It also presents a dual motion‑model tracker to handle non‑linear perspective changes and a semantic attribute classification pipeline that estimates occlusion, readability, sign embeddedness, and road relevance. Evaluated on a dataset covering more than 60 countries and 2,500 hours of driving, the system achieves an Object Miss Ratio of 0.49% across 221,068 sequences, indicating strong global generalization for autonomous driving.
The paper introduces a dataset‑centric benchmark for deep learning approaches to grape leaf disease classification and detection. It evaluates publicly available datasets on disease categories, annotations, acquisition conditions, and class distributions, and tests representative models across image‑level classification, region‑level classification, and object detection. Results reveal high accuracy on controlled datasets but significant performance drops on heterogeneous, real‑world data, especially in cross‑dataset transfer and object detection tasks.
The paper investigates how visual presentation affects vision‑language models (VLMs) on the SPaRC spatial planning benchmark. By adding lightweight input‑side scaffolds that keep the visual modality but make spatial structure clearer, the authors achieve up to a 34.0‑percentage‑point accuracy boost across multiple VLMs, and an additional 4.6 points when combined with GRPO training. Analyses reveal that these improvements stem mainly from reduced grounding errors, while rule‑based reasoning remains difficult, highlighting visual presentation as a key determinant of whether VLM benchmarks test grounded perception, downstream reasoning, or both.
The paper introduces ANT, a test‑time adaptation framework that improves prostate cancer detection in micro‑ultrasound by performing a segmentation‑guided adaptation. ANT aligns a pretrained detection encoder to the target domain’s prostate anatomy using pseudo‑masks from a frozen segmentation network, thereby correcting domain‑specific feature drift while preserving cancer‑discriminative features. In a leave‑one‑center‑out evaluation, ANT raises mean AUC by 2.9% at the biopsy‑core level and 3.6% at the patient level compared to no adaptation, outperforming existing TTA baselines.
The paper introduces CFM, a language‑aligned concept foundation model for vision that generates fine‑grained, human‑interpretable concepts with spatial grounding. By pairing CFM with a strong semantic foundation model, it provides explanations for downstream tasks such as classification, segmentation, and captioning. The authors also analyze local co‑occurrence of concepts to define relationships, improving concept naming and yielding richer explanations while maintaining competitive performance.
The study introduces a new optical coherence tomography (OCT) dataset of chronically implanted guinea pig cochleae to quantify intracochlear fibrosis, a factor that can impair residual hearing in cochlear implant users. Researchers applied several semantic segmentation models, finding that a modified UNET architecture (2D‑OCT‑UNET) on upscaled images achieved the best performance in identifying fibrosis and related structures. The dataset and code are publicly available for further research.
AT‑ViT is a dual‑branch Vision Transformer that processes both raw herbarium scans and their segmentation masks through a multi‑scale, multi‑view cross‑attention fusion. It uses a mask‑guided patch weighting scheme to emphasize plant regions and suppress background artifacts, thereby encouraging plant‑centric representations. In trait classification tasks such as leaf base shape and thorns, AT‑ViT consistently outperforms baselines, improves spatial attention grounding (IoU_p +15.66 to +18.03 pp, IoU_b –27.92 to –31.02 pp), and shows greater robustness to synthetic background perturbations, surpassing ResNet101 by up to +32.32 accuracy points and CrossViT by up to +5.07 points. whyItMatters":"The model addresses shortcut learning caused by background cues in herbarium images, leading to more accurate and interpretable plant trait recognition."
The article reviews how transformer models, retrieval‑grounded pipelines, and large language models are reshaping hadith computational science. It critically evaluates existing literature, highlighting uneven progress: expanded data resources and mature segmentation tasks, yet persistent issues such as narrow corpora, weak benchmark comparability, and limited reproducibility. The authors argue that hadith computation should be viewed as an evidence‑infrastructure problem requiring knowledge integration, provenance, and expert supervision, and they propose a research agenda to strengthen the field’s methodological rigor.
arXiv:2608.21300v1 Announce Type: new Abstract: Foundation models for medical image segmentation, like prompt-based MedSAM, generalize well across domains and modalities, often in zero or few-shot se...
arXiv:2404.09431v3 Announce Type: replace Abstract: Pseudo-LiDAR has become a promising paradigm for monocular 3D object detection by transforming monocular images into point cloud representations th...
arXiv:2608.18012v2 Announce Type: replace Abstract: One of the most common reasons for anterior cruciate ligament (ACL) reconstruction failure is femoral tunnel malpositioning (ACL footprint center a...