Pith. sign in

REVIEW 28 cited by

YOLOv10: Real-Time End-to-End Object Detection

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.14458 v2 pith:MD6D7EGB submitted 2024-05-23 cs.CV

Ao Wang , Hui Chen , Lihao Liu , Kai Chen , Zijia Lin , Jungong Han , Guiguang Ding This is my paper
classification cs.CV
keywords yolosperformancedetectionmodelcomputationalefficiencyend-to-endlatency
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Over the past years, YOLOs have emerged as the predominant paradigm in the field of real-time object detection owing to their effective balance between computational cost and detection performance. Researchers have explored the architectural designs, optimization objectives, data augmentation strategies, and others for YOLOs, achieving notable progress. However, the reliance on the non-maximum suppression (NMS) for post-processing hampers the end-to-end deployment of YOLOs and adversely impacts the inference latency. Besides, the design of various components in YOLOs lacks the comprehensive and thorough inspection, resulting in noticeable computational redundancy and limiting the model's capability. It renders the suboptimal efficiency, along with considerable potential for performance improvements. In this work, we aim to further advance the performance-efficiency boundary of YOLOs from both the post-processing and model architecture. To this end, we first present the consistent dual assignments for NMS-free training of YOLOs, which brings competitive performance and low inference latency simultaneously. Moreover, we introduce the holistic efficiency-accuracy driven model design strategy for YOLOs. We comprehensively optimize various components of YOLOs from both efficiency and accuracy perspectives, which greatly reduces the computational overhead and enhances the capability. The outcome of our effort is a new generation of YOLO series for real-time end-to-end object detection, dubbed YOLOv10. Extensive experiments show that YOLOv10 achieves state-of-the-art performance and efficiency across various model scales. For example, our YOLOv10-S is 1.8$\times$ faster than RT-DETR-R18 under the similar AP on COCO, meanwhile enjoying 2.8$\times$ smaller number of parameters and FLOPs. Compared with YOLOv9-C, YOLOv10-B has 46\% less latency and 25\% fewer parameters for the same performance.

Discussion (0). Sign in to comment.

Forward citations

Cited by 28 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 1,045 citations worldwide. Full citation record

  1. ShrinkBox: Backdoor Attack on Object Detection to Disrupt Collision Avoidance in Machine Learning-based Advanced Driver Assistance Systems

    cs.CV 2025-07 conditional novelty 7.0 of 10

    ShrinkBox is a backdoor attack that makes object detectors output smaller bounding boxes on triggered images, preserving mAP while tripling distance-estimation error.

  2. Dual-Correlation Hypergraph Network for Unaligned RGBT Video Object Detection and A Large-scale Benchmark

    cs.CV 2026-07 conditional novelty 6.0 of 10

    DHNet with patch alignment and dual hypergraph fusion reaches SOTA RGBT video object detection on VT-VOD50 and the new large-scale DVT-VOD1000 benchmark.

  3. Tracing 3D Anatomy in 2D Strokes: A Multi-Stage Projection Driven Approach to Cervical Spine Fracture Identification

    cs.CV 2026-01 conditional novelty 6.0 of 10

    Cervical spine fractures can be detected from vertebra volumes reconstructed from 2D CT projections at F1 68.15 (vertebra) and 82.26 (patient), matching a reproduced 3D winner within about 1 point.

  4. MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimation

    cs.CV 2025-08 conditional novelty 6.0 of 10

    MV-SSM models the joint spatial sequence with state-space layers plus projective attention and bidirectional token scanning, improving multi-view 3D pose accuracy and generalization to unseen camera setups.

  5. Recognizing Actions from Robotic View for Natural Human-Robot Interaction

    cs.CV 2025-07 conditional novelty 6.0 of 10

    ACTIVE is a new RGB+LiDAR action recognition dataset from moving robotic viewpoints at 3-50m, and ACTIVE-PC improves point cloud action recognition on it to 60.10% Top-1 accuracy.

  6. An Uncertainty-aware DETR Enhancement Framework for Object Detection

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Adding a Gaussian-box Gromov-Wasserstein loss and Bayes-risk-based refinement to DETR detectors improves their AP on COCO and leukocyte datasets while producing localization uncertainty estimates.

  7. Rainbow Artifacts from Electromagnetic Signal Injection Attacks on Image Sensors

    cs.CR 2025-07 conditional novelty 6.0 of 10

    Tuned electromagnetic interference at an 84.68 MHz carrier paints rainbow color bands onto CMOS camera images and degrades YOLO, Faster R-CNN, and DETR object detection.

  8. A Comprehensive Dataset for Underground Miner Detection in Diverse Scenario

    cs.CV 2025-06 conditional novelty 6.0 of 10

    The authors present a 7,049-image thermal dataset of underground miners in five posture classes and report transfer-learning baselines, with best mAP50 around 80%.

  9. Benchmarking Deep Learning Models for Dense Event Classification of Offshore Wind Infrastructure in Sentinel-1 Time Series

    cs.LG 2026-08 conditional novelty 5.0 of 10

    A supervised BiLSTM improves dense event classification of offshore wind Sentinel-1 time series over the rule-based baseline (AUCEditSim 0.7853 to 0.8509), and the resulting labels expose regional deployment dynamics.

  10. An Analysis of Layer-Freezing Strategies for Enhanced Transfer Learning in YOLO Architectures

    cs.CV 2025-09 conditional novelty 5.0 of 10

    Freezing the first four blocks or the whole backbone of YOLOv8/YOLOv10 can match or beat full fine-tuning while using less GPU memory, but aggressive freezing fails on heavily augmented single-class data.

  11. Image Quality Enhancement and Detection of Small and Dense Objects in Industrial Recycling Processes

    cs.CV 2025-09 conditional novelty 5.0 of 10

    A new industrial recycling dataset and benchmark show YOLOv8-x and YOLO11-x as top detectors, and a small denoising and super-resolution network (DDSRNet) improves noisy image quality.

  12. Multi-Representation Adapter with Neural Architecture Search for Efficient Range-Doppler Radar Object Detection

    cs.CV 2025-09 conditional novelty 5.0 of 10

    A YOLOv8-based detector with a multi-representation heatmap/grayscale adapter branch and one-shot NAS reports new SOTA mAP@50 of 71.9 (RADDet) and 57.1 (CARRADA).

  13. DOMR: Establishing Cross-View Segmentation via Dense Object Matching

    cs.CV 2025-08 conditional novelty 5.0 of 10

    DOMR jointly matches and refines multiple object masks across ego and exo views, reaching 49.7% and 55.2% mean IoU on Ego-Exo4D.

  14. PhaseNAS: Language-Model Driven Architecture Search with Dynamic Phase Adaptation

    cs.LG 2025-07 reject novelty 5.0 of 10

    PhaseNAS uses dynamic small-to-large LLM switching and a template language to search neural architectures, claiming better accuracy and lower search cost on NAS-Bench-Macro, CIFAR, and COCO.

  15. FishDet-M: A Unified Large-Scale Benchmark for Robust Fish Detection and CLIP-Guided Model Selection in Diverse Aquatic Visual Domains

    cs.CV 2025-07 conditional novelty 5.0 of 10

    FishDet-M merges 13 fish detection datasets into a 105,556-image COCO-style benchmark and benchmarks 28 detectors, with YOLO12x reaching the highest mAP of 0.491.

  16. An Intelligent-Cloud Edge Multimodal Interaction System for Robots

    cs.RO 2026-07 conditional novelty 4.0 of 10

    A cloud-edge robot system combined a CBAM/DIoU-enhanced YOLO11n gesture detector with LLM/VLM agents, reporting 95–98.9% precision and 82–95% task success on small, validation-based evaluations.

  17. Thresholded Cross-Attention for Reliable Intensity-Chromaticity Fusion in Low-Light Image Enhancement

    cs.CV 2026-07 conditional novelty 4.0 of 10

    TCA-Net improves low-light image enhancement by replacing fixed-quota Top-K attention with an adaptive confidence threshold for intensity-chromaticity fusion in the HVI color space.

  18. A Multi-Task Deep Learning Framework for Real-Time Intelligent Video Surveillance with Temporal Event Validation

    cs.CV 2026-07 conditional novelty 4.0 of 10

    Temporal multi-frame validation around concurrent CV detectors cuts fire false alarms from 52% to 4% and lifts video plate exact-match from 66.7% to 81.8% at real-time latency on one commodity GPU.

  19. CollabOD: Collaborative Multi-Backbone with Cross-scale Vision for UAV Small Object Detection

    cs.CV 2026-03 conditional novelty 4.0 of 10

    CollabOD improves UAV small-object detection via dual-path detail preservation, dense aggregation, bilateral reweighting, and a reparameterized detail-aware head, reporting 52.4 AP50 on VisDrone at 65.5 GFLOPs.

  20. YOLO-Master: MOE-Accelerated with Specialized Transformers for Enhanced Real-time Detection

    cs.CV 2025-12 conditional novelty 4.0 of 10

    YOLO-Master inserts a sparse Mixture-of-Experts block into a YOLO backbone, reporting 42.4% COCO AP at 1.62 ms, +0.8 AP and 18% faster than YOLOv13-N.

  21. PaveSync: A Unified and Comprehensive Dataset for Pavement Distress Analysis and Classification

    cs.CV 2025-12 conditional novelty 4.0 of 10

    PaveSync merges existing pavement imagery into a standardized 52,747-image, 13-class detection dataset and benchmarks seven object detectors on it.

  22. A Survey on the Applications of Generative Artificial Intelligence in Automated Driving Systems Test Scenario Generation Methods

    cs.SE 2025-12 reject novelty 4.0 of 10

    A literature survey of scenario-generation methods for ADS testing that adds an unvalidated AII/RAS/OCS metric suite and ODD-difficulty schema, undermined by inconsistent calculations in the worked examples.

  23. YOLO-ROC: A High-Precision and Ultra-Lightweight Model for Real-Time Road Damage Detection

    cs.CV 2025-07 conditional novelty 4.0 of 10

    YOLO-ROC reports 67.6% mAP50 on RDD2022_China_Drone with 0.89M parameters, a 1.4-point gain over YOLOv8n.

  24. PlantSAM: An Object Detection-Driven Segmentation Pipeline for Herbarium Specimens

    cs.CV 2025-07 conditional novelty 4.0 of 10

    PlantSAM, a pipeline that uses YOLOv10 bounding boxes to prompt SAM2, segments plants from herbarium backgrounds with IoU 0.94 and improves trait classification accuracy by up to 4.36%.

  25. CSDN: A Context-Gated Self-Adaptive Detection Network for Real-Time Object Detection

    cs.CV 2025-06 conditional novelty 4.0 of 10

    A plug-and-play Transformer detection head with gated block, neighbor, and deformable attention improves YOLO-family COCO AP by 0.7 to 1.0 points.

  26. UniDet-D: A Unified Dynamic Spectral Attention Model for Object Detection under Adverse Weathers

    cs.CV 2025-06 conditional novelty 4.0 of 10

    A unified detection-plus-restoration model with learnable DCT frequency gating reports modest mAP gains over multi-task baselines across rain, fog, snow, and low-light tests, with weaker quantitative support for unsee...

  27. Policy-Driven Transfer Learning in Resource-Limited Animal Monitoring

    cs.CV 2025-09 reject novelty 3.0 of 10

    Using a UCB bandit selection, the framework identifies RTDETRx as the best pre-trained model for animal detection with F1=0.718, versus 0.690 for exhaustive selection, while running fewer models per image.

  28. SOD-YOLO: Enhancing YOLO-Based Detection of Small Objects in UAV Imagery

    cs.CV 2025-07 conditional novelty 3.0 of 10

    A YOLOv8 variant using ASF feature fusion, a P2 detection head, and Soft-NMS reports 36.1% higher mAP50:95 on VisDrone2019-DET than its YOLOv8-m baseline.

Pith tools