Real-time object detector RT-DETR beats YOLO in accuracy and speed by efficient multi-scaling

--

Real-time object detector RT-DETR beats YOLO in accuracy and speed by efficient multi-scaling

DETRs Beat YOLOs on Real-time Object Detection
arXiv paper abstract https://arxiv.org/abs/2304.08069
arXiv PDF paper https://arxiv.org/pdf/2304.08069.pdf

… end-to-end transformer-based detectors (DETRs) have … remarkable performance. However … high computational cost … preventing … no post-processing, such as non-maximum suppression (NMS).

… first analyze the influence of NMS in modern real-time object detectors on inference speed …

To avoid … delay … by NMS, … propose a Real-Time DEtection TRansformer (RT-DETR), the first real-time end-to-end object detector

… design an efficient hybrid encoder to efficiently process multi-scale features by decoupling the intra-scale interaction and cross-scale fusion, and propose IoU-aware query selection to improve the initialization of object queries.

… detector supports flexibly adjustment of the inference speed by using different decoder layers without the need for retraining, which facilitates the practical application of real-time object detectors.

… RT-DETR-L … on T4 GPU … outperforming all YOLO detectors of the same scale in both speed and accuracy …

Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website

LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b

Photo by Collab Media on Unsplash

--

--

AI News Clips by Morris Lee: News to help your R&D
AI News Clips by Morris Lee: News to help your R&D

Written by AI News Clips by Morris Lee: News to help your R&D

A computer vision consultant in artificial intelligence and related hitech technologies 37+ years. Am innovator with 66+ patents and ready to help a firm's R&D.

No responses yet