RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lv, Wenyu, Zhao, Yian, Chang, Qinyao, Huang, Kui, Wang, Guanzhong, Liu, Yi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916335033778176
author Lv, Wenyu
Zhao, Yian
Chang, Qinyao
Huang, Kui
Wang, Guanzhong
Liu, Yi
author_facet Lv, Wenyu
Zhao, Yian
Chang, Qinyao
Huang, Kui
Wang, Guanzhong
Liu, Yi
contents In this report, we present RT-DETRv2, an improved Real-Time DEtection TRansformer (RT-DETR). RT-DETRv2 builds upon the previous state-of-the-art real-time detector, RT-DETR, and opens up a set of bag-of-freebies for flexibility and practicality, as well as optimizing the training strategy to achieve enhanced performance. To improve the flexibility, we suggest setting a distinct number of sampling points for features at different scales in the deformable attention to achieve selective multi-scale feature extraction by the decoder. To enhance practicality, we propose an optional discrete sampling operator to replace the grid_sample operator that is specific to RT-DETR compared to YOLOs. This removes the deployment constraints typically associated with DETRs. For the training strategy, we propose dynamic data augmentation and scale-adaptive hyperparameters customization to improve performance without loss of speed. Source code and pre-trained models will be available at https://github.com/lyuwenyu/RT-DETR.
format Preprint
id arxiv_https___arxiv_org_abs_2407_17140
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer
Lv, Wenyu
Zhao, Yian
Chang, Qinyao
Huang, Kui
Wang, Guanzhong
Liu, Yi
Computer Vision and Pattern Recognition
In this report, we present RT-DETRv2, an improved Real-Time DEtection TRansformer (RT-DETR). RT-DETRv2 builds upon the previous state-of-the-art real-time detector, RT-DETR, and opens up a set of bag-of-freebies for flexibility and practicality, as well as optimizing the training strategy to achieve enhanced performance. To improve the flexibility, we suggest setting a distinct number of sampling points for features at different scales in the deformable attention to achieve selective multi-scale feature extraction by the decoder. To enhance practicality, we propose an optional discrete sampling operator to replace the grid_sample operator that is specific to RT-DETR compared to YOLOs. This removes the deployment constraints typically associated with DETRs. For the training strategy, we propose dynamic data augmentation and scale-adaptive hyperparameters customization to improve performance without loss of speed. Source code and pre-trained models will be available at https://github.com/lyuwenyu/RT-DETR.
title RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.17140