-
Hello, I noticed that in the article, the eval result for qpic in V-COCO is not 58.8 as mentioned in the original paper but 61.8. Is there a specific reason for this difference, and could it be relate…
-
### Model/Dataset/Scheduler description
这个模型是DETR系列，是DINO的延续，首次统一了检测和分割任务，并且在检测和分割各自的任务上指标都是sota
既有mask2former的分割优点，又有DINO检测的优点
他们的代码也是基于mmlab的框架，已开源
强烈推荐
### Open source status
- [X] The model …
-
Thank you very much for your contribution!
I want to know why the Query POS Embedding in Semantically Aligner has the operation of "*0.5" before it passes through Sin POS Embedding.
https://github.c…
-
I'm trying to do some error analysis. While looking through predictions in results.pkl I noticed there are overlapping bounding boxes with IoU larger than the threshold in the config file. I wasn't ab…
-
@thibo73800 To reproduce, you can launch a train (detr for example) with this modification if you have only one GPU :
Replace :
`accelerator=None if torch.cuda.device_count()
-
I have run the DETR successfully and get some results but the issue is interpretation? what is maxdets and whit is meaning of area?
-
Thank you for sharing the code. I want to ask you how to set the epoch50 for your training? I verified that 12 and 24 are OK, but the results of 50 rounds of Deformable Detr are almost the same as tho…
-
- examine plant specimen page structure
- multiple labels (original label, [update label](https://scc-ondemand1.bu.edu/pun/sys/files/fs/projectnb/sparkgrp/ml-herbarium-grp/ml-herbarium-data-spring20…
-
It would be more useful to have a object detection model since most of the computer vision modules are directed to NVIDIA.
-
作者您好，我在2080ti上测输入640×640，bs=1，pytorch版本RT-DETR R50的推理时间是25ms，YOLOv6-L的推理时间是17毫秒，论文里两者的FPS几乎一样，请问推理时间为什么差这么多？
LM986 updated
6 months ago