Detection, segmentation, depth and recognition research, plus the vision backbones that keep displacing the last generation.
arXiv:2508.13680v5 Announce Type: replace-cross Abstract: We introduce VMMU, a Vietnamese Multitask Multimodal Understanding and Reasoning Benchmark designed to evaluate how vision-language models (V...
Recent advances in video object segmentation with Multimodal Large Language Model (MLLM) reasoning have demonstrated the effectiveness of using a single textual token, such as SEG, to predict segmenta...
Textual descriptions can reduce ambiguity in medical image segmentation by specifying the finding and location to be delineated. Existing text-guided methods mainly improve where image and language fe...
Although hyperspectral images (HSIs) provide rich spectral-spatial information, accurate pixel-level classification remains challenging because of spectral-spatial heterogeneity and complex spatial st...
With the development of applications like autonomous driving, object detection has gained significant attention, while also highlighting critical vulnerabilities like backdoor attacks that severely co...
Segmentation-based multi-object tracking (MOT) with foundation video models such as SAM2 offers strong localization quality, yet remains fragile in crowded, real-world scenes. In detector-prompted SAM...
Video understanding is usually benchmarked on curated, single-actor, or professionally filmed clips, and a strong score there is routinely read as evidence a model is robust enough for deployment. Ama...
Methods operating on Vision Transformer (ViT) feature spaces typically rely on Euclidean distance or cosine similarity. This assumes that every direction is equally meaningful, but there is no reason...
4D radar complements dense image semantics with long-range geometry and radial motion, but existing radar--camera detectors largely solve \emph{where} to align the modalities while leaving \emph{wheth...
Stereo visual SLAM systems built on local descriptors suffer from semantic ambiguity, instance-level confusion, and independently moving objects, each corrupting data association and accumulating as t...
Most computer-vision systems organize visual input toward a predefined interpretation, such as semantic categories, prompted regions, learned object-like representations, or a single spatial partition...
arXiv:2609.26756v1 Announce Type: new Abstract: X-ray is medicine's most widely used imaging modality, yet remains among its least quantitative. Unlike volumetric modalities like CT or MRI, X-ray col...
SAM‑V is a geometry‑aware extension of the Segment Anything Model (SAM) that integrates 3D priors from a feed‑forward geometry model (VGGT) into 2D segmentation. It uses a prompt‑fusion mechanism to combine sparse SAM prompts with view‑specific camera tokens and local VGGT features, enabling a mask decoder that attends to both dense 2D and 3D cues. The resulting end‑to‑end system produces consistent multi‑view instance segmentation in a single forward pass, achieving significant gains on the IGGT 3D tracking benchmark without offline mask matching or explicit 3D reconstruction.
RootQuantV2 adapts a frozen DINOv3 Vision Transformer to predict root length and surface area directly from minirhizotron images, eliminating the need for manually traced masks. By training only 11.9 M parameters (3.78 % of the model), it achieves R² values of 0.950 for length and 0.930 for area, improving RMSE by 24.3 % and 20.7 % over the previous RootQuant CNN approach. The method repurposes existing numeric archives of root traits for high‑throughput, automated phenotyping in field‑grown crops.
The paper introduces Gated Token Recurrence (GTR), a softmax‑free recurrent vision backbone that replaces global softmax attention with gated linear attention, alternating scan directions, and enhanced SwiGLU blocks. GTR is distilled from a DINOv3 teacher using only final‑layer patch‑token alignment, and achieves strong performance on COCO object detection (58.9 box AP) with very low latency (1.908 ms on an RTX 4090). The backbone also transfers to multiple dense prediction tasks and runs efficiently on edge hardware via a specialized CUDA operator and TensorRT deployment.
The paper introduces a lightweight CPU-based extension to GROBID that uses layout-guided masking to identify figure, table, and paratext regions in scientific PDFs. By routing tokens to specialized GROBID models or discarding them, the method improves structural accuracy on PMC corpora and enhances figure caption recovery. It also achieves competitive table detection and body‑text precision compared to vision‑based GPU parsers while operating entirely on CPU and costing significantly less.
The study investigates how the choice of annotation used to set a segmentation threshold influences retinal vessel segmentation performance. Using all 28 CHASE DB1 images and two human observers, the authors fit random forests and Extra Trees models, then compare five threshold policies—including fixed, observer‑tuned, mean‑observer, and maximin tuning—on the same score maps. Results show that maximin tuning alters thresholds in most fits but yields negligible changes in worst‑observer Dice scores, suggesting no accuracy advantage in this cohort.
The paper introduces MedIDL, a Medical Imaging Disentanglement Learning framework that separates disease-related features from confounding covariates and individual variability in medical images. It achieves this by projecting image features into three orthogonal latent spaces—disease classification, covariate alignment, and a Gaussian head for individual variation—using specialized disentanglement heads. Across seven diverse imaging datasets, MedIDL surpasses state‑of‑the‑art supervised and self‑supervised methods in classification accuracy, and its latent representations and gradient‑based visualizations align with known clinical patterns.
The study investigates how contour initialization and stopping tolerance jointly influence active‑contour segmentation performance on dermoscopic images. Using scikit‑image’s Chan‑Vese implementation on a resized ISIC 2017 dataset, the authors compare Otsu, checkerboard, disk, and Otsu‑initialized contours under default and tighter level‑set tolerances. Results show that Otsu initialization improves mean Dice scores at default tolerance but the advantage diminishes with tighter tolerance, and that evaluation metrics can vary across lesion size strata, underscoring the need to report initial fields, tolerances, and iteration counts.
C2FXNet is a unified object detection framework designed for adverse weather conditions. It uses a dual-level guidance mechanism: a Multi-step Reasoning Router (MRR) for coarse scene reasoning and a Fine Scene Refinement (FSR) module for fine-grained semantic adjustment. A Scene-aware Mixture-of-Experts (SMoE) dynamically combines scene-specific experts, enabling robust detection across foggy, dark, and clear scenes without scene-specific training.