BCNet

BCNet copied to clipboard

Where is the config file "fcos_imprv_R_50_FPN_1x.yaml"?

1

The shell all.sh used that config file, but I did not find it, there is only "fcos_imprv_R_50_FPN.yaml" in the directory "configs/fcos", how could I get it? BTW, thanks for this...

qgq99

Sorry, it is difficult for me to follow the methodology. As the Bayesian model is learned from the images without occluded objects, what is the role of shape modeling/prior during...

FedericaGielen

Link to the "converted mask annotation" is not working

8

Hello, the link to the "converted mask annotation" is not working. Could you update the link or is there another way I can download it? Thank you.

codefortheplanet

Thank you for your great work! I am trying to train a model to perform Amodal Instance Segmentation and input my own images to see the prediction results of hidden...

masa-suke

How do Networks predict amodal masks

1

Hi, I have read your paper. And i have noticed that in the Experiments part, you evaluate the performance of BCNets on amodal dataset like cocoa. With figure4 i know...

staple-pi

Request for BCNet + Faster R-CNN weights for evaluation

3

Hi, thank you for sharing your work. Would you be able to share your trained weights for the BCNet + Faster R-CNN (with ResNet-50 backbone) model trained on COCO train2017,...

levan92

No code for BCNet on Faster/Mask RCNN

1

Hi, I realise that BCNet implementation in Mask RCNN is also not given in this repo. Will you be able to provide them for purpose of results reproduction? Thank you!

levan92

Hello, I followed the instructions to install the environment and ran into this problem: RuntimeError: No such operator torchvision::nms I am using 2080Ti, here is my environment information： _libgcc_mutex 0.1...

ZNan-Chen

what is the inference time or fps achieved on your hardware ?

1

could someone let me know how the model performs on your hardware, it would really help if you could let me know the inference time or fps achieved with the...

riririririi

train issues

2

WARNING [10/15 22:01:14 fvcore.common.checkpoint]: The checkpoint state_dict contains keys that are not used by the model: fc1000.{bias, weight} stem.conv1.bias [10/15 22:01:14 d2.engine.train_loop]: Starting training from iteration 0 ERROR [10/15 22:01:15...

C-Ll-l