← Back to all news

Computer vision

Detection, segmentation, depth and recognition research, plus the vision backbones that keep displacing the last generation.

3,375 stories · RSS feed

arXiv Computer Vision
Sep 23

Interpretable AI plus Handheld, Portable Retinal Photographs: A Low-Cost Glaucoma Screening Solution for West Africa

The study presents an interpretable AI framework for glaucoma screening using low-cost handheld retinal cameras in a West African population. Trained on 681 participants, the system achieved high performance across vessel segmentation, cup/disc segmentation, and optic nerve head feature detection, with classification AUCs of 0.85 for the handheld device and 0.93 for a tabletop camera. The model provides confidence scores and visual explanations to aid clinical interpretation.

By Charis Y. N. Chiang, Tarela Sarimiye, Adeyinka Ashaye, Martin Buist, Michael A. Hauser, Olusola Olawoye, Micha\"el J. A. Girard
computer-visionfine-tuning
More like this →
arXiv Computer Vision
Sep 23

SPEANet: Structural Prior Enhanced Attention Network for Parameter-Efficient Remote Sensing Object Detection

SPEANet is a parameter‑efficient backbone for remote sensing object detection that incorporates fixed structural operators to enhance contour and frequency responses. It uses stage‑specific prior extraction and context‑conditioned response modulation, assigning smoothed contour and directional modeling to shallow features and a compact approximation‑detail interaction to deeper stages. Experiments on five benchmarks, including DOTA‑v1.0, DOTA‑v1.5, and DIOR‑R, show that SPEANet achieves high mAP scores (up to 78.55%) with only 23.0 M total parameters, of which the backbone accounts for 5.97 M.

By Wei Lu, Junjie Li, Feifei Sang, Si-Bao Chen
computer-visionbenchmarks
More like this →
arXiv Computer Vision
Sep 23

Match One, Learn with Graph: One-to-Graph Query Collaboration with Backward Sharing for Object Detection

The paper introduces BS‑O2G, a plug‑in that constructs a sparse prediction‑aware graph from decoded features, boxes, and class distributions to enable one‑to‑graph query collaboration in Detection Transformers. It uses One‑to‑Graph (O2G) calibration to propagate messages forward and Backward Sharing (BS) to route gradients backward, preserving the original one‑to‑one matcher and positive labels. Experiments on various DETR models, backbones, COCO, and CrowdHuman datasets demonstrate consistent performance gains, faster convergence, and minimal additional parameters or FLOPs.

By Wenxiao Fan, Jingling Fu, Luohang Liu, Lichen Ma, Yu He, Zhiyang Yu, Weishan Bi, Junshi Huang, Yan Li, Gu Simiu, Kan Li
llmscomputer-vision
More like this →
arXiv Computer Vision
Sep 23

Semantically-Guided Domain Randomization for Industrial Object Detection in Low-Image-Budget Regimes

Semantically-Guided Domain Randomization (S‑GDR) is an annotation‑free pipeline that uses vision‑language model captioning of a small real reference set, diffusion‑based background synthesis, and mask‑based object composition to generate synthetic training data. In a high‑mix, low‑volume automotive detection benchmark, S‑GDR achieves a mAP50‑95 of 0.739 with only 200 synthetic images, outperforming a domain‑randomized render baseline and several other synthetic data methods under the same budget. These results suggest S‑GDR is a viable alternative for training visual perception systems when annotation, energy, and time resources are severely limited.

By Jose Moises Araya-Martinez, Gautham Mohan, Jens Lambrecht
llmsdiffusioncomputer-visionmultimodalbenchmarks
More like this →
arXiv Computer Vision
Sep 23

Vision Foundation Models with Synthetic-Only Training for Monocular Spacecraft Pose Estimation

The paper presents a new monocular spacecraft pose estimation model that achieves the lowest reported mean rotation errors on the SPEED+ lightbox and sunlamp test sets. By replacing smaller encoders with a large self‑supervised ViT foundation model (DINOv3) and scaling up to 840 M parameters, the authors improve accuracy from 300 M to 840 M parameters without saturation. The 840 M model also runs on a Jetson Orin NX 16 GB with 133.8 ms per crop and 32.0 W power draw, demonstrating embedded inference feasibility while training solely on synthetic data.

By John Church, Vazghen Nikolian
computer-visionfine-tuning
More like this →
arXiv Computer Vision
Sep 23

Laryngeal Structure Segmentation in High-Speed Videoendoscopy Using Deep Learning

The study presents deep learning‑based segmentation of laryngeal structures in high‑speed videoendoscopy (HSV) data, including connected speech. Using U‑Net models, the authors trained networks to detect aryepiglottic folds, arytenoid cartilages, vocal folds, epiglottis, and glottal area from HSV recordings of both sustained vowel phonation and connected speech in normal and disordered voices. After applying noise removal and histogram equalization, the networks achieved overall accuracies above 95%, demonstrating reliable automated analysis of laryngeal dynamics.

By Sardar Nafis Bin Ali, Mohsen Zayernouri, Dimitar D. Deliyski, Maryam Naghibolhosseini
computer-vision
More like this →
arXiv Computer Vision
Sep 23

GAD-MambaUNet: Direction-Group Mamba with Gradient-Adaptive DINOv3 Distillation for Lightweight Medical Image Segmentation

The paper introduces GAD-MambaUNet, a lightweight medical image segmentation network that integrates efficient local modeling, Direction-Group Graph Selective Scan (DG‑GSS) for structured information exchange, and training‑time supervision from a frozen DINOv3 teacher with Gradient‑Adaptive Distillation. GAD‑MambaUNet demonstrates a strong accuracy‑efficiency trade‑off compared to other lightweight and general segmentation methods, and ablation studies confirm the benefits of DG‑GSS and DINOv3‑GAD supervision. Future work aims to refine teacher‑student alignment and apply the framework to multi‑class and multi‑modal medical segmentation tasks.

By Fang Wang, Huitao Li, Wenhan Chao, Zheng Zhuo, Xinxin Yang
computer-visionefficiencymultimodalbenchmarkssafety
More like this →
arXiv Computer Vision
Sep 23

TEMPURA: Temporal Event Masked Prediction and Understanding for Reasoning in Action

arXiv:2505.01583v2 Announce Type: replace Abstract: Understanding causal event relationships and achieving fine-grained temporal grounding in videos remain challenging for vision-language models (VLM...

By Jen-Hao Cheng, Yi-Hao Peng, Huapeng Zhou, Vivian Wang, Huayu Wang, Hsiang-Wei Huang, Wenhao Chai, Hou-I Liu, Kuang-Ming Chen, Cheng-Yen Yang, Yi-Ling Chen, Vibhav Vineet, Qin Cai, Jenq-Neng Hwang
llmscomputer-visionmultimodalbenchmarks
More like this →