Closed ymzis69 closed 1 month ago
Actually, the DN-DETR + Group DETR method improves mAP when training is done solely on the COCO dataset without pre-training. However, when a pre-training followed by a fine-tuning process is used, the performance gain from the DeNoising training method is minimal. Therefore, to simplify the approach, we only used Group DETR training method.
Thank you for your reply. I wish you success in your studies and work.
Thank you for your amazing work. I have a question. In your Group DETR paper, it is noted that the DN-DETR + Group DETR method improves the mAP. Why did you only use the Group DETR training method in LW-DETR but not the DeNoising training method? Have you tried adding the DeNoising method in LW-DETR experiments? Looking forward to your response.