Closed xmu-xiaoma666 closed 2 years ago
I am still running experiments.
Is there any update of the performance?
By adding new augmentation, stable-pix2seq could achieve 34mAP at 218 epoches
By adding new augmentation, stable-pix2seq could achieve 34mAP at 218 epoches
The paper reports 45.0mAP?
The reported performance is 43.0 mAP. The paper lack details about the implementation. We are trying our best to replicate the results reported in the paper. As the training process is slow, it may take several months to replicate the results. We will update the results on this GitHub once we have progress,
256 epoch 36mAP. Beam search will addition improve 1-2 mAP. The estimated mAP is 37-38mAP at 256 epoch.
Will you release the beam search code?
All code including beam search and data augmentation will be released.
Does Pix2Seq converge slower than DETR?
Due to different learning schedules, it's hard to compare the convergence speed between DETR and Pix2Seq.
We released the pretrained model at https://github.com/gaopengcuhk/Pretrained-Pix2Seq.
Thank you for your implementation!!! Can you share Performance Difference Compared With Original Paper and checkpoint?