From 702455586a48955b96256ff44a1abeb7e4429485 Mon Sep 17 00:00:00 2001 From: hjin2902 Date: Thu, 1 Jul 2021 22:23:58 +0800 Subject: [PATCH 1/4] Bump to v0.15.0 --- README.md | 4 ++-- README_zh-CN.md | 4 ++-- docs/changelog.md | 42 ++++++++++++++++++++++++++++++++++++++++- docs/getting_started.md | 15 ++++++--------- mmdet3d/version.py | 2 +- 5 files changed, 52 insertions(+), 15 deletions(-) diff --git a/README.md b/README.md index 766b50273..27ae10d60 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,7 @@ [![license](https://img.shields.io/github/license/open-mmlab/mmdetection3d.svg)](https://github.com/open-mmlab/mmdetection3d/blob/master/LICENSE) -**News**: We released the codebase v0.14.0. +**News**: We released the codebase v0.15.0. In the recent [nuScenes 3D detection challenge](https://www.nuscenes.org/object-detection?externalData=all&mapData=all&modalities=Any) of the 5th AI Driving Olympics in NeurIPS 2020, we obtained the best PKL award and the second runner-up by multi-modality entry, and the best vision-only results. @@ -62,7 +62,7 @@ This project is released under the [Apache 2.0 license](LICENSE). ## Changelog -v0.14.0 was released in 1/6/2021. +v0.15.0 was released in 1/7/2021. Please refer to [changelog.md](docs/changelog.md) for details and release history. ## Benchmark and model zoo diff --git a/README_zh-CN.md b/README_zh-CN.md index 4214e5e6a..4b60de0e4 100644 --- a/README_zh-CN.md +++ b/README_zh-CN.md @@ -8,7 +8,7 @@ [![license](https://img.shields.io/github/license/open-mmlab/mmdetection3d.svg)](https://github.com/open-mmlab/mmdetection3d/blob/master/LICENSE) -**新闻**: 我们发布了版本v0.14.0. +**新闻**: 我们发布了版本v0.15.0. 在第三届[ nuScenes 3D 检测挑战赛](https://www.nuscenes.org/object-detection?externalData=all&mapData=all&modalities=Any)（第五届 AI Driving Olympics, NeurIPS 2020）中，我们获得了最佳 PKL 奖、第三名和最好的纯视觉的结果，相关的代码和模型将会在不久后发布。 @@ -62,7 +62,7 @@ MMDetection3D 是一个基于 PyTorch 的目标检测开源工具箱, 下一代 ## 更新日志 -最新的版本 v0.14.0 在 2021.06.01发布。 +最新的版本 v0.15.0 在 2021.07.01发布。 如果想了解更多版本更新细节和历史信息，请阅读[更新日志](docs/changelog.md)。 ## 基准测试和模型库 diff --git a/docs/changelog.md b/docs/changelog.md index 461566025..8af8fc8af 100644 --- a/docs/changelog.md +++ b/docs/changelog.md @@ -1,5 +1,45 @@ ## Changelog +### v0.15.0 (1/7/2021) + +#### Highlights + +- Support [PAConv](https://arxiv.org/abs/2103.14635) +- Support monocular/multi-view 3D detector [ImVoxelNet] (https://arxiv.org/abs/2106.01178) on KITTI +- Support Transformer-based 3D detection method [Group-Free-3D](https://arxiv.org/abs/2104.00678) on ScanNet +- Add documentation for tasks including LiDAR-based 3D detection and point-based 3D semantic segmentation +- Add dataset documents like ScanNet +- Refine docs for tutorials, Quick Run and Useful Tools + +#### New Features + +- Support Group-Free-3D on ScanNet (#539) +- Support PAConv modules(#598, #599) +- Support ImVoxelNet on KITTI (#627, #654) + +#### Improvements + +- Add unit tests for pipeline functions `LoadImageFromFileMono3D`, `ObjectNameFilter` and `ObjectRangeFilter` (#615) +- Enhance [IndoorPatchPointSample](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/pipelines/transforms_3d.py) (#617) +- Refactoring model initialization methods based MMCV (#622) +- Add Chinese docs (#629) +- Add documentation for LiDAR-based 3D detection (#642) +- Unifying intrinsic and extrinsic matrices for all datasets (#653) +- Add documentation for point-based 3D semantic segmentation (#663) +- Add documentation of ScanNet for 3D detection (#664) +- Refine docs for tutorials (#666) +- Refine docs for Quick Run and Useful Tools (#686) + + +#### Bug Fixes + +- Fix the bugs of [BackgroundPointsFilter](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/pipelines/transforms_3d.py) (#609) +- Fix LoadMultiViewImageFromFiles to be compatible with DefaultFormatBundle (#611) +- Fix the potential bug in [analyze_logs](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/analysis_tools/analyze_logs.py) (#634) +- Fix test command in docs and made some refinements (#635) +- Fix wrong config paths in unit tests (#641) + + ### v0.14.0 (1/6/2021) #### Highlights @@ -20,7 +60,7 @@ - Support visualization of detection results and dataset browse for nuScenes Mono-3D dataset (#542, #582) - Support faster implementation of KNN (#586) - Support RegNetX models on Lyft dataset (#589) -- Remove a useless parameter [label_weight] from segmentation datasets including [Custom3DSegDataset], [ScanNetSegDataset] and [S3DISSegDataset] (#607) +- Remove a useless parameter `label_weight` from segmentation datasets including `Custom3DSegDataset`, `ScanNetSegDataset` and `S3DISSegDataset` (#607) #### Bug Fixes - Fix a corrupted lidar data file in Lyft dataset in [data_preparation](https://github.com/open-mmlab/mmdetection3d/tree/master/docs/data_preparation.md) (#546) diff --git a/docs/getting_started.md b/docs/getting_started.md index ae4e9aa59..539a780fb 100644 --- a/docs/getting_started.md +++ b/docs/getting_started.md @@ -12,7 +12,8 @@ The required versions of MMCV, MMDetection and MMSegmentation for different vers | MMDetection3D version | MMDetection version | MMSegmentation version | MMCV version | |:-------------------:|:-------------------:|:-------------------:|:-------------------:| -| master | mmdet>=2.10.0, <=2.11.0| mmseg==0.14.0 | mmcv-full>=1.3.1, <=1.4| +| master | mmdet>=2.14.0| mmseg==0.14.1 | mmcv-full>=1.3.1, <=1.4| +| 0.15.0 | mmdet>=2.14.0| mmseg==0.14.1 | mmcv-full>=1.3.1, <=1.4| | 0.14.0 | mmdet>=2.10.0, <=2.11.0| mmseg==0.14.0 | mmcv-full>=1.3.1, <=1.4| | 0.13.0 | mmdet>=2.10.0, <=2.11.0| Not required | mmcv-full>=1.2.4, <=1.4| | 0.12.0 | mmdet>=2.5.0, <=2.11.0 | Not required | mmcv-full>=1.2.4, <=1.4| @@ -94,12 +95,8 @@ pip install mmcv-full **d. Install [MMDetection](https://github.com/open-mmlab/mmdetection).** -Note: - -MMDetection3D v0.14.0 is only compatible with MMDetection version `mmdet>=2.10.0, <=2.11.0`. The future versions will only support `mmdet>=2.12.0` since v0.15.0 (to be released in July). - ```shell -pip install mmdet==2.11.0 +pip install mmdet==2.14.0 ``` Optionally, you could also build MMDetection from source in case you want to modify the code: @@ -107,7 +104,7 @@ Optionally, you could also build MMDetection from source in case you want to mod ```shell git clone https://github.com/open-mmlab/mmdetection.git cd mmdetection -git checkout v2.11.0 # switch to v2.11.0 branch +git checkout v2.14.0 # switch to v2.14.0 branch pip install -r requirements/build.txt pip install -v -e . # or "python setup.py develop" ``` @@ -115,7 +112,7 @@ pip install -v -e . # or "python setup.py develop" **e. Install [MMSegmentation](https://github.com/open-mmlab/mmsegmentation).** ```shell -pip install mmsegmentation==0.14.0 +pip install mmsegmentation==0.14.1 ``` Optionally, you could also build MMSegmentation from source in case you want to modify the code: @@ -123,7 +120,7 @@ Optionally, you could also build MMSegmentation from source in case you want to ```shell git clone https://github.com/open-mmlab/mmsegmentation.git cd mmsegmentation -git checkout v0.14.0 # switch to v0.14.0 branch +git checkout v0.14.1 # switch to v0.14.1 branch pip install -e . # or "python setup.py develop" ``` diff --git a/mmdet3d/version.py b/mmdet3d/version.py index 1afa91b49..410f0ef19 100644 --- a/mmdet3d/version.py +++ b/mmdet3d/version.py @@ -1,6 +1,6 @@ # Copyright (c) Open-MMLab. All rights reserved. -__version__ = '0.14.0' +__version__ = '0.15.0' short_version = __version__ From 39567e6799561768605fbdc34297061115c5e01d Mon Sep 17 00:00:00 2001 From: hjin2902 Date: Fri, 2 Jul 2021 11:51:47 +0800 Subject: [PATCH 2/4] fix typo, specific bugfix, install version --- docs/changelog.md | 20 ++++++++++---------- docs/getting_started.md | 4 ++-- requirements/mminstall.txt | 6 +++--- 3 files changed, 15 insertions(+), 15 deletions(-) diff --git a/docs/changelog.md b/docs/changelog.md index 8af8fc8af..db5c24e21 100644 --- a/docs/changelog.md +++ b/docs/changelog.md @@ -5,38 +5,38 @@ #### Highlights - Support [PAConv](https://arxiv.org/abs/2103.14635) -- Support monocular/multi-view 3D detector [ImVoxelNet] (https://arxiv.org/abs/2106.01178) on KITTI +- Support monocular/multi-view 3D detector [ImVoxelNet](https://arxiv.org/abs/2106.01178) on KITTI - Support Transformer-based 3D detection method [Group-Free-3D](https://arxiv.org/abs/2104.00678) on ScanNet -- Add documentation for tasks including LiDAR-based 3D detection and point-based 3D semantic segmentation +- Add documentation for tasks including LiDAR-based 3D detection, vision-only 3D detection and point-based 3D semantic segmentation - Add dataset documents like ScanNet -- Refine docs for tutorials, Quick Run and Useful Tools #### New Features - Support Group-Free-3D on ScanNet (#539) -- Support PAConv modules(#598, #599) +- Support PAConv modules (#598, #599) - Support ImVoxelNet on KITTI (#627, #654) #### Improvements - Add unit tests for pipeline functions `LoadImageFromFileMono3D`, `ObjectNameFilter` and `ObjectRangeFilter` (#615) - Enhance [IndoorPatchPointSample](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/pipelines/transforms_3d.py) (#617) -- Refactoring model initialization methods based MMCV (#622) +- Refactor model initialization methods based MMCV (#622) - Add Chinese docs (#629) - Add documentation for LiDAR-based 3D detection (#642) -- Unifying intrinsic and extrinsic matrices for all datasets (#653) +- Unify intrinsic and extrinsic matrices for all datasets (#653) - Add documentation for point-based 3D semantic segmentation (#663) - Add documentation of ScanNet for 3D detection (#664) - Refine docs for tutorials (#666) +- Add documentation for vision-only 3D detection (#669) - Refine docs for Quick Run and Useful Tools (#686) #### Bug Fixes -- Fix the bugs of [BackgroundPointsFilter](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/pipelines/transforms_3d.py) (#609) -- Fix LoadMultiViewImageFromFiles to be compatible with DefaultFormatBundle (#611) -- Fix the potential bug in [analyze_logs](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/analysis_tools/analyze_logs.py) (#634) -- Fix test command in docs and made some refinements (#635) +- Fix the bug of [BackgroundPointsFilter](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/pipelines/transforms_3d.py) using the bottom center of ground truth (#609) +- Fix LoadMultiViewImageFromFiles to get list format for DefaultFormatBundle (#611) +- Fix the potential bug in [analyze_logs](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/analysis_tools/analyze_logs.py) when the training resumes from a checkpoint or is stopped before evaluation (#634) +- Fix test commands in docs and make some refinements (#635) - Fix wrong config paths in unit tests (#641) diff --git a/docs/getting_started.md b/docs/getting_started.md index 539a780fb..31ca52a0a 100644 --- a/docs/getting_started.md +++ b/docs/getting_started.md @@ -12,8 +12,8 @@ The required versions of MMCV, MMDetection and MMSegmentation for different vers | MMDetection3D version | MMDetection version | MMSegmentation version | MMCV version | |:-------------------:|:-------------------:|:-------------------:|:-------------------:| -| master | mmdet>=2.14.0| mmseg==0.14.1 | mmcv-full>=1.3.1, <=1.4| -| 0.15.0 | mmdet>=2.14.0| mmseg==0.14.1 | mmcv-full>=1.3.1, <=1.4| +| master | mmdet>=2.14.0, <=3.0.0| mmseg>=0.14.1, <=1.0.0 | mmcv-full>=1.3.8, <=1.4| +| 0.15.0 | mmdet>=2.14.0, <=3.0.0| mmseg>=0.14.1, <=1.0.0 | mmcv-full>=1.3.8, <=1.4| | 0.14.0 | mmdet>=2.10.0, <=2.11.0| mmseg==0.14.0 | mmcv-full>=1.3.1, <=1.4| | 0.13.0 | mmdet>=2.10.0, <=2.11.0| Not required | mmcv-full>=1.2.4, <=1.4| | 0.12.0 | mmdet>=2.5.0, <=2.11.0 | Not required | mmcv-full>=1.2.4, <=1.4| diff --git a/requirements/mminstall.txt b/requirements/mminstall.txt index 3e107b56b..634bef8af 100644 --- a/requirements/mminstall.txt +++ b/requirements/mminstall.txt @@ -1,3 +1,3 @@ -mmcv-full>=1.3.1,<=1.4.0 -mmdet>=2.10.0,<=2.11.0 -mmsegmentation>=0.14.0,<=0.14.0 +mmcv-full>=1.3.8,<=1.4.0 +mmdet>=2.14.0,<=3.0.0 +mmsegmentation>=0.14.1,<=1.0.0 From f416f9753178fea7a2cfcf7e1cb9eaf218a12a5a Mon Sep 17 00:00:00 2001 From: hjin2902 Date: Fri, 2 Jul 2021 13:27:22 +0800 Subject: [PATCH 3/4] fix LoadMultiViewImageFromFiles --- docs/changelog.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/changelog.md b/docs/changelog.md index db5c24e21..3ffa973a2 100644 --- a/docs/changelog.md +++ b/docs/changelog.md @@ -34,7 +34,7 @@ #### Bug Fixes - Fix the bug of [BackgroundPointsFilter](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/pipelines/transforms_3d.py) using the bottom center of ground truth (#609) -- Fix LoadMultiViewImageFromFiles to get list format for DefaultFormatBundle (#611) +- Fix [LoadMultiViewImageFromFiles](https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/pipelines/loading.py) to unravel stacked multi-view images to list to be consistent with DefaultFormatBundle (#611) - Fix the potential bug in [analyze_logs](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/analysis_tools/analyze_logs.py) when the training resumes from a checkpoint or is stopped before evaluation (#634) - Fix test commands in docs and make some refinements (#635) - Fix wrong config paths in unit tests (#641) From 4d230f62fe5fb5e8641b1825ae8b81c3633c96a2 Mon Sep 17 00:00:00 2001 From: hjin2902 Date: Fri, 2 Jul 2021 14:21:29 +0800 Subject: [PATCH 4/4] compatibility about mmcv version --- docs/changelog.md | 4 ++++ docs/compatibility.md | 4 ++++ 2 files changed, 8 insertions(+) diff --git a/docs/changelog.md b/docs/changelog.md index 3ffa973a2..0d6e05e3d 100644 --- a/docs/changelog.md +++ b/docs/changelog.md @@ -2,6 +2,10 @@ ### v0.15.0 (1/7/2021) +#### Compatibility + +In order to fix the problem that the priority of EvalHook is too low, all hook priorities have been re-adjusted in 1.3.8, so MMDetection 2.14.0 needs to rely on the latest MMCV 1.3.8 version. For related information, please refer to [#1120](https://github.com/open-mmlab/mmcv/pull/1120), for related issues, please refer to [#5343](https://github.com/open-mmlab/mmdetection/issues/5343). + #### Highlights - Support [PAConv](https://arxiv.org/abs/2103.14635) diff --git a/docs/compatibility.md b/docs/compatibility.md index 26e0c74d8..cca2522bc 100644 --- a/docs/compatibility.md +++ b/docs/compatibility.md @@ -4,6 +4,10 @@ This document provides detailed descriptions of the BC-breaking changes in MMDet ## MMDetection3D 0.15.0 +### MMCV Version + +In order to fix the problem that the priority of EvalHook is too low, all hook priorities have been re-adjusted in 1.3.8, so MMDetection 2.14.0 needs to rely on the latest MMCV 1.3.8 version. For related information, please refer to [#1120](https://github.com/open-mmlab/mmcv/pull/1120), for related issues, please refer to [#5343](https://github.com/open-mmlab/mmdetection/issues/5343). + ### Unified parameter initialization To unify the parameter initialization in OpenMMLab projects, MMCV supports `BaseModule` that accepts `init_cfg` to allow the modules' parameters initialized in a flexible and unified manner. Now the users need to explicitly call `model.init_weights()` in the training script to initialize the model (as in [here](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/train.py#L183), previously this was handled by the detector. Please refer to PR #622 for details.