GitLab

- fix 'invalid configuration argument' error triggered by empty point input. test cases covering similar situations are added to test_dynamic_scatter.py as well.
trivial changes:
- switch to using torch::unique_dim to generate reduce mapping instead of calculating it from scratch.

* support nuimages visualization

* recover config

* support nuimages visualization

* support batchsize > 1

* add comments

* remove max_num_point in ScanNet data preprocessing

* add config file for ScanNet semantic segmentation dataset

* modify NormalizePointsColor in pipeline

* add visualization function for semantic segmentation

* add ignore_index to semantic segmentation visualization function

* add ignore_index to semantic segmentation evaluation function

* fix ignore_index bug in semantic segmentation evaluation function

* add test function to check ignore_index assignment in PointSegClassMapping

* fix slicing bug in BasePoints class and add unittest

* add IndoorPatchPointSample class for indoor semantic segmentation data loading and add unittest

* modify LoadPointsFromFile class and its unittest to support point color loading

* fix data path in unittest

* add setter function for coord and attributes of BasePoint and modify unittest

* modify color normalization operation to work on BasePoint class

* add unittest for ScanNet semantic segmentation data loading pipeline

* fix ignore_index bug in seg_eval function

* add ScanNet semantic segmentation dataset and unittest

* modify config file for ScanNet semantic segmentation

* fix visualization function and modify unittest

* fix a typo in seg_eval.py

* raise exception when semantic mask is not provided in train/eval data loading

* support custom computation of label weight for loss calculation

* modify seg_eval function to be more efficient

* fix small bugs & change variable names for clarity & add more cases to unittest

* move room index resampling and label weight computation to data pre-processing

* add option allowing user to determine whether to sub-sample point clouds

* fix typos & change .format to f-string & fix link in comment

* save all visualizations into .obj format for consistency

* infer num_classes from label2cat in eval_seg function

* add pre-computed room index and label weight for ScanNet dataset

* replace .ply with .obj in unittests and documents

* add TODO in case data is on ceph

* add base dataset for all semantic segmentation tasks & add ScanNet dataset inheriting from base dataset

* rename class for consistency

* fix minor typos in comment

* move Custom3DSegDataset to a new file

* modify BasePoint setter function to enable attribute adding

* add unittest for NormalizePointsColor and fix small bugs

* fix unittest for BasePoints

* modify ScanNet data pre-processing scripts

* change ignore_idx to -1 in seg_eval function

* remove sliding inference from PatchSample function and modify unittest

* remove PatchSample from scannet seg test_pipeline

* fix pt1.8 issues

* fix pt1.8 issues

* modify vote_head to support 3dssd

* delete .keys()

* add 3dssd unittest

* 3dssd->ssd3d

* sunrgbd doc update and change in image path

* rephrase

* fix compilation error in pytorch 1.7

* add pt1.7 build

* Update build.yml

* fix link to convert_votenet_checkpoints.py

* remove redundant space in docstring

* fix typos

* fix link and latex citation in README

* fix link error for publish_model.py

* fix link error for regnet2mmdet.py

* fix link error for files under tools/misc/ folder

* fix link error for files under tools/analysis_tools/ folder

* fix latex citation in README_zh-CN.md

* fix typo and a link to mmdet

* support nuscenes dataset in demo

* add convert_SyncBN in __init__

* fix meshlab visualization bug

* modify meshlab unittest

* add docstring

* add empty line in docstring
Co-authored-by: xiliu8006 <xiliu800@gmail.com>

* Added image loading in SUNRGB-D dataset (#195)

* image loading

* format and docstring fix

* removed irrelevant files

* removed irrelevant files

* load image only if modality is pc+img

* added modality like nuscenes

* Added imvotenet image branch pretrain (#217)

* image loading

* naive commit

* format and docstring fix

* removed irrelevant files

* removed irrelevant files

* load image only if modality is pc+img

* added modality like nuscenes

* pretrain_2d_model

* finetune sunrgbd

* finetune sunrgbd

* deleted unused code

* fixed a bug

* resolve conflict

* update config file

* fix docstring and configs

* integrated vote fusion

* coords transform and unit test

* Update docstring

* refactor and add unit test

* fix bug caused by mmcv upgrade; delete pdb breakpoint

* add point fusion unittest

* remove unused file

* fix typos

* updates

* add assertion info

* update

* add unittest

* add vote fusion unittest

* add vote fusion unittest

* [Refactor] VoteNet refactor (#322)

* votenet refactor

* remove file

* minor update

* docstring

* initial update of imvotenet

* [Feature] Support vote fusion (#297)

* integrated vote fusion

* coords transform and unit test

* Update docstring

* refactor and add unit test

* add point fusion unittest

* remove unused file

* updates

* add assertion info

* update

* add unittest

* add vote fusion unittest

* add vote fusion unittest

* minor update

* docstring

* change np ops to torch

* refactor test

* update

* refactor of image mlp and np random ops to torch

* add docstring

* add config and mod dataset

* fix bugs

* add_comments

* fix bugs

* fix_bug

* fix bug

* fix bug

* fix bug

* fix bug

* final fix

* fix bug

* ?

* add docstring

* move train/test cfg

* change img mlp default param

* rename config

* minor mod

* change config name

* move train/test cfg

* some fixes and 2d utils

* fix config name

* fix config override issue

* config simplify & reformat

* explicitly set eval mode->override train()

* add fix_img_branch to config

* remove set_img_branch_eval_mode

* temporal fix, change calibs to calib

* more docstring and view/reshape, expand/repeat change

* complete imvotenet docstring

* fix docstring

* add config and some minor fix

* rename config
Co-authored-by: ZwwWayne <wayne.zw@outlook.com>

* Avoid 4 dims points

* add assert prinf

* modify print info

* support knn gpu op

* made it more robust and fixed comments

* bugfix and add warning in docstring

* support knn gpu op

* made it more robust and fixed comments

* add ini

* add semantickitti_dataset

* add test semantickitti_dataset

* delete last line in test_semmaticdataset

* add test data

* change_names

* load_labels dytype

* change_name

* numpy

* name

* dtype string

* minor issue-string

* seg_3d_dtype

* add_fast hist

* add per_class iou

* add seg_eval

* add unitest

* minor error

* minor error2

* minor error3

* Update README.md

* Update README_zh-CN.md

* Update getting_started.md

* Update version.py

* Update changelog.md

* Update changelog.md

* Update changelog.md

[Fix] fix a bug that may cause compilation failure of dynamic voxelization when using GPUs with compute capability lower than 6.x (#326)

* fix a bug that may cause compilation failure of dynamic voxelization when using gpus with compute capability lower than 6.x
fix imperfection kernel code that may unintentionally discard valid points when input points count is larger than 50000 * 512 (nearly impossible though).

* Modified scatter_points_cuda.cu to ensure backward compatibility with PyTorch1.5 on CUDA9.0

* fix the issue of DynamicScatter gradient check failure by explicit mark non-floating-point tensor as non-differentiable.

* fix nonzero warning with as_tuple

* fix nonzero warning with as_tuple #320

* Reduce the open3d dependency

* Reduce the open3d dependency

* Reduce dependency of open3d if open3d is not used

* remove the warning in show_result

* a faster & more memory-efficient implementation of DynamicScatter

* fix format issues and add pytest skip code for tests on machines without cuda support

* some trivial changes:

decrease the number of kernel threads per block to 512, to enable inference on GPUs with computing capability lower than 2.0

change the backpropagation behavior of max-reduction. when there are multiple points shares the same maximum feature value, only the first point (with lowest row index) among them is chosen to propagate the output gradient back. before this change, all points with the same maximum feature value can propagate the output gradient back. this change makes the max-reduction behaves in consistence with torch.max. this change may cause gradcheck failure in test_dynamic_scatter.py. please do not worry about it because torch.max fails the gradcheck too.

* fix typo
Co-authored-by: zhanggefan <1152009@tongji.edu.cn>

* add h3d backbone

* add h3d backbone

* add h3dnet

* modify scannet config

* fix bugs for proposal refine

* fix bugs for test backbone

* add primitive head test

* modify h3dhead

* modify h3d head

* update loss weight config

* fix bugs for h3d head loss

* modify h3d head get targets function

* update h3dnet base config

* modify weighted loss

* Revert "Merge branch 'h3d_u2' into 'master'"

This reverts merge request !5

* add open3d visual

* modify docstring

* use bbox and coord convert

* modify docstring

* fix bugs for unittest

* fix bugs for unittest

* modify doc

* add visualize script and modify docs

* modify docs

* fix bugs

* fix bug caused by mmcv upgrade; delete pdb breakpoint

* fix typos

* use torch.cuda api

* update unittest for points_in_bbox_gpu and points_in_boxes_batch

* Added comments for explanation

* Move train_cfg/test_cfg to model

* Move train_cfg/test_cfg to model

* Move train_cfg/test_cfg to model

* Move train_cfg/test_cfg to model

* Move train_cfg/test_cfg to model

* Move train_cfg/test_cfg to model

* Move train_cfg/test_cfg to model

* Move train_cfg and test_cfg into model

* modify centerpoint configs

* Modify docs

* modify build_detector

* modify test_config_build_detector

* modify build_detector parameters

* Adopt the same strategy in build_detector

* fix bug caused by mmcv upgrade; delete pdb breakpoint

* fix typos

* Update README.md

* Update getting_started.md

* Update version.py

* Update changelog.md

* Update numpy requirements

* Update runtime.txt

* Merge minor items

* Add support for SemanticKITTI dataset

* add ini

* add semantickitti_dataset

* add test semantickitti_dataset

* delete last line in test_semmaticdataset

* add test data

* change_names

* load_labels dytype

* change_name

* numpy

* name

* dtype string

* minor issue-string

* seg_3d_dtype

* fixed the bug about miss text files on kitti testing

* add bbox2result_kitti unit test

* add bbox2result_kitti unit test

* add exist_submission_file unit test

* deleted debug code and convert{} to dict()

* add _get_kitti_dataset_config()

* delete print

* Fix bug when num_features != 4

* add voxelization unittest

* fixed CI without GPU

* use the numpy version to test the CUDA version
Co-authored-by: Guanghui Ren(任广辉) <sundrops.ren@gmail.com>

The tuple unpacked into three arguments when four were given when
running VoteNet on sunrgbd

* add h3d backbone

* add h3d backbone

* add h3dnet

* modify scannet config

* fix bugs for proposal refine

* fix bugs for test backbone

* add primitive head test

* modify h3dhead

* modify h3d head

* update loss weight config

* fix bugs for h3d head loss

* modify h3d head get targets function

* update h3dnet base config

* modify weighted loss

* Revert "Merge branch 'h3d_u2' into 'master'"

This reverts merge request !5

* fix bugs for empty scene

* modify doc

* modify filter empty scene

* Update README.md

* Update version.py

* Update getting_started.md

* Update __init__.py

* Update changelog.md

* Update README.md

* Update news for nuScenes detection challenge

* Update README to be compatible with refactored doc

* Update an installation subtitle

* rephrase
Co-authored-by: ZwwWayne <wayne.zw@outlook.com>

* Fix an old bug in box_np_ops, not used or influencing results

* Update a minor filter criterion in KITTI evaluation

* Update the fixed criterion, not influencing KITTI results

* fix indoor_eval in case of less classes in prediction

* fix style for formatters