GitLab

* add the code of generating cam_sync_labels in waymo dataset

* fix key error in waymo converation

* fix waymo convert bug and refactor parse_ann_info in WaymoDataset

* add image_path key in waymo sweeps

* polish code

* add sweeps in data_prefix in waymo config

* fix some bugs in waymo

* resolve some comments and fix configs of pointpillars

* add 2 TODOs

* fix lint
Co-authored-by: lianqing <lianqing1997@gmail.com>

* update evaluation metric to support waymo cam only evaluation

* add transformation for bev detection

* add multiview dfm

* support multiview detection in datasets with transformation, dfm model and metric

* remove deprecated config and update doc string

* remove file_client_args=file_client_args and update docstr

* add doc string and remove pdb

* fix the doc string of voxel fusion

* add doc string

* remove lidar2img

* add doc string

* update doc string

* support waymo dataset for replace_ceph and modify path of pkl in config

* update evaluation metrics; and the config for waymo solution

* fix the index error in waymo metric and add lidar2img utils function

* replace __call_ to transform

* fix doc string

* rename configs

* update the config name

* update the lidar2cam calib in waymo data creater
Co-authored-by: lianqing <lianqing1997@gmail.com>
Co-authored-by: Tai-Wang <tab_wang@outlook.com>

* update vis doc

* support bev visualization and update proj_3d visualization

* fix bugs

* fix comments and add visualizer in dataset config

* fix comments

* add draw_points_on_image sample

* fix ci

* remove mkdir_or_exist

* update vis doc

* fix mono3d related bugs

* update kitti-mono3d script

* update mono3d task

* update resize3d and randomresize3d

* fix

* update dataset converter script

* fix part of comments

* unify the task name in datasets and visualization

* fix comments

* rename 3d to lidar_det

* fix ci

* change boxlist to boxtype

* change default value ot lidar_det

* fix bugs

* fix ci

* remove spconv in optional.txt

* update pre-commit-hook config

* fix typos in new pre-commit-hook

* fix nuscenes training bug

* update kitti_dataset docstring

* update nuscenes_dataset docstring

* fix lint and modifty docstring

* fix the doc

* resolve comments

* fix doct

* update customize

* update customize dataset

* update doc

* update doc

* fix comments

* fix comments

* update dataset transforms

* update dbsampler docstring and add typehints

* add type hints and fix potential point sample bug

* fix lint

* fix

* fix

* add deploy.yaml

* fix

* fix

* fic

* fix

* fix

* fix

* fix

* fix

* fix

* fix bug

* fix comments

* fix comments

* fix

* fix

* Minor fix
Co-authored-by: Tai-Wang <tab_wang@outlook.com>

* modify ci config

* minor changes

* min python==3.7.4 & update install docs

* fix data_preprocessor

* fix data_preprocessor

* install waymo-open-dataset-tf-2-6-0

* update docs

* add coco ann for mono3d demo and fix inference

* add demo pkl

* update docs

* fix ci of dataset and visualizer by regenerate pkl

* decrease channels in ut for reducing cuda memory

* still use sample_id in the kitti submitted files

* minor changes

* use waymo-open-dataset-tf-2-6-0

* use waymo-open-dataset-tf-2-5-0 to be compabible with python3.6

* use waymo-open-dataset-tf-2-4-0 to be compabible with python3.6

* no dependency about waymo-open-dataset

* Revert "no dependency about waymo-open-dataset"

This reverts commit 581d4eefe26851a295b3190bcfa73d6fb2ba54aa.

* not install waymo-open-dataset for ci

* max python version is 3.8 because of trimesh

* Revert "max python version is 2.8 because of trimesh"

This reverts commit 6d8a1851b22a2059e7de75b576c2b7f633ad7cfa.

* fix networkx in ci

* update docstring of data_preprocessor

* modify version of typingextensions

* update benchmark config

* fix docstrings

* update docstrings

* update benchmark hook

* fix nuscenes datasets

* fix fcos3d bugs

* ImVoxelNet inherits from Base3DDetector

* fix typo sample_id

* replace add_pred_to_datasample with add_pred_to_datasample

* refactor imvoxelnet config

* feat(local_visualizer.py): add check dir exist

* fix(local_visualizer.py): use mmengine API

* format(local_visualizer.py): remove useless

* style(local_visualizer.py): format

* add deploy.yaml

* fix ut

* fix update data script description

* force the mask as bool

* fix config name

* Revert "fix config name"

This reverts commit 99fc6445d9a49824b61692bcdd374eabd0ef3200.

* remove type convert

* minor changes

* fix docs

* add deploy.yaml

* fix inference

* add deploy.yaml

* add auto_scale_lr in cfg

* fix cfg

* update repo version

* update

* update version.py

* fix version

* fix liar_det3d

* fix ch liar_det3d

* fix en and ch liar_seg and add formart results

* resolve the comments

* fix the nuscence

* fix scannet

* fix waymo det

* fix docs

* fix the doc

* fix the command line

* fix the command line

* fix nuscence

* fix nuscence

* resolve the comments

* resp;ve

* Minor fix

* Minor fix
Co-authored-by: Tai-Wang <tab_wang@outlook.com>

* fix bugs

* fix bugs 2

* fix seg demo

* update use collate data in mmengine

* update ceph in command

* update comments

* update

* fix lint

* fix readme

* fix cn

* fix dcos

* fix comments

* Update release news

* Update README_zh-CN.md

* Update README.md

* fix docs
Co-authored-by: Tai-Wang <tab_wang@outlook.com>

* fix custom runtime

* fix custom_model

* fix cfg

* fix comments

* fix comments

* fix comments