The recently proposed Detection Transformer (DETR) model successfully applies Transformer to objects detection and achieves comparable performance with two-stage object detection frameworks, such as Faster-RCNN. However, DETR suffers from its slow convergence. Training DETR from scratch needs 500 epochs to achieve a high accuracy. To accelerate its convergence, we propose a simple yet effective scheme for improving the DETR framework, namely Spatially Modulated Co-Attention (SMCA) mechanism. The core idea of SMCA is to conduct location-aware co-attention in DETR by constraining co-attention responses to be high near initially estimated bounding box locations. Our proposed SMCA increases DETR's convergence speed by replacing the original co-attention mechanism in the decoder while keeping other operations in DETR unchanged. Furthermore, by integrating multi-head and scale-selection attention designs into SMCA, our fully-fledged SMCA can achieve better performance compared to DETR with a dilated convolution-based backbone (45.6 mAP at 108 epochs vs. 43.3 mAP at 500 epochs). We perform extensive ablation studies on COCO dataset to validate SMCA. Code is released at https://github.com/gaopengcuhk/SMCA-DETR .
@article{arxiv.2108.02404,
title = {Fast Convergence of DETR with Spatially Modulated Co-Attention},
author = {Peng Gao and Minghang Zheng and Xiaogang Wang and Jifeng Dai and Hongsheng Li},
journal= {arXiv preprint arXiv:2108.02404},
year = {2021}
}
Comments
Accepted by ICCV2021. arXiv admin note: substantial text overlap with arXiv:2101.07448