HuangJunJie2017/BEVDet issues and pull requests

#28 - Run mono_det_demo.py problem

Issue - State: closed - Opened by Jumponthemoon over 2 years ago - 2 comments
Labels: enhancement

#27 - Problem when testing

Issue - State: closed - Opened by galaxyGGG over 2 years ago - 1 comment

#26 - fix bug in post-processing

Pull Request - State: closed - Opened by Yzichen over 2 years ago

#25 - I found the gt_bbox sending to the forward_train function is flipped compared to the gt_bbox in the outputs of csgbdatasets. Where is the code for the flip above.

Issue - State: closed - Opened by drilistbox over 2 years ago - 6 comments

#24 - GPU memory consumption

Issue - State: closed - Opened by XuyangBai over 2 years ago - 1 comment

#23 - questions about DETR3D inference speed

Issue - State: closed - Opened by czy341181 over 2 years ago - 3 comments

#22 - Can't inference by the master code

Issue - State: closed - Opened by pycoco over 2 years ago - 7 comments

#21 - Collecting reproducing results of BEVDet-Tiny

Issue - State: closed - Opened by HuangJunJie2017 over 2 years ago - 29 comments
Labels: help wanted

#20 - Questions about retraining

Issue - State: closed - Opened by gaojunbin over 2 years ago - 4 comments

#19 - Question about BEV Augmentation

Issue - State: closed - Opened by kaixinbear over 2 years ago - 6 comments

#18 - The mASE and mAOE is too large

Issue - State: closed - Opened by KeyuLi over 2 years ago - 10 comments

#17 - How to use your pretrained model to do inference on nuScenes mini dataset?

Issue - State: closed - Opened by cherry-yh over 2 years ago - 5 comments

#16 - After retraining, I found that the projection results always shift forward. If the official model is used, it is normal. Has the author encountered it?

Issue - State: closed - Opened by abing222 over 2 years ago - 16 comments

#15 - Evaluation results on nuScenes using pretrained weight

Issue - State: closed - Opened by BoLang615 over 2 years ago - 5 comments

#14 - Question about intermediate BEV feature of the previous frame

Issue - State: closed - Opened by ged0606 over 2 years ago - 2 comments

#13 - NuScenesDataset: init() got an unexpected keyword argument 'img_info_prototype'

Issue - State: closed - Opened by Egozjuer over 2 years ago - 11 comments

#12 - NuScenesDataset: init() got an unexpected keyword argument 'img_info_prototype'

Issue - State: closed - Opened by Egozjuer over 2 years ago - 1 comment

#11 - Question about the inference speed.

Issue - State: closed - Opened by exiawsh over 2 years ago - 1 comment

#10 - Visualize output of model

Issue - State: closed - Opened by idontlikelongname over 2 years ago - 5 comments

#9 - I have some confusion in reproduce Bev4D with resnet101 config.

Issue - State: closed - Opened by Crownwzt over 2 years ago - 1 comment

#8 - TypeError: save_for_backward can only save variables, but argument 1 is of type tuple

Issue - State: closed - Opened by drilistbox over 2 years ago - 16 comments

#7 - AttributeError: 'LoadMultiViewImageFromFiles_BEVDet' object has no attribute 'camera_model_consistent'

Issue - State: closed - Opened by drilistbox over 2 years ago - 2 comments

#6 - share detection results for a val set in .JSON file #4

Issue - State: closed - Opened by sunwell1994 almost 3 years ago - 1 comment

#5 - modle training details for nuscenes test split

Issue - State: closed - Opened by jianingwangind almost 3 years ago - 3 comments

#4 - When will the code be released? Thank you!

Issue - State: closed - Opened by weixin00 about 3 years ago - 5 comments

#3 - Questions about model training

Issue - State: closed - Opened by jianingwangind about 3 years ago - 1 comment

#2 - The output format

Issue - State: closed - Opened by tonyyang1995 about 3 years ago - 1 comment

#1 - 请问什么时候开源代码？

Issue - State: closed - Opened by totem367 about 3 years ago - 4 comments

Ecosyste.ms: Issues

GitHub / HuangJunJie2017/BEVDet issues and pull requests