fix(tensorrt_yolox): add run length encoding for sematic segmentation mask #7905

badai-nguyen · 2024-07-09T00:29:16Z

Description

Since transfering the raw image of sematic segmentation mask between Perception ECU where tensorrt running and Main ECU causes the bandwidth issue and increases tensorrt_yolox node processing_time.
Using JPEG compressed method from image_transport can reduce bandwidth but pixel value might be changed during decompression that lead the changed of semantic segmentation especially along the boundary of segment.
Using lossless compression (PNG) of image_transport improve the bandwidth and processing_time but not good as run_length method. The reason is that semantic mask of road scene normally has large segment with the same class which is very effective for run_length encoder
Comparison of bandwidth and processing time between image_transport lossless compression (PNG) with the runlength when runing multi-header model on ROSCUBE and transfer result to Main ECU:

raw iamge	'image_transport`'s PNG` png_level: 1`	'image_transport`'s PNG` png_level: 9`	implemented 'run_length'
Average : 91.5351569282137 ,std: 0.5338352012729306	Average: 54.99859166666666 ,std: 1.066178226938265	Average: 135.7696519721578 ,std: 10.981303755107556	Average: 49.28519 ,std: 0.7785528951629853

Releated PR: fix(image_projection_based_fusion): add run length decoding for segmentation_pointcloud_fusion #7909
Related Issue: Add a new pointcloud filter based on image semantic segmentation #5594

How was this PR tested?

Running ros2 launch autoware_tensorrt_yolox yolox_s_plus_opt.launch.xml on Embedded Ecu and confirmed that processing time is unchanged when sematic segmentation mask topic topic is subscribed from outside of Embedded ECU.
- Tested reults on Embedded Robotic Controller Powered by NVIDIA® Jetson AGX Xavier ™ Module RQX-58G with preprocess_on_gpu: true:
- OS version:

cat /etc/nv_tegra_release
-> R32 (release), REVISION: 6.1, GCID: 27863751, BOARD: t186ref, EABI: aarch64, DATE: Mon Jul 26 19:36:31 UTC 2021

Confirmed that publishing_time is significantly decreased, so that the overal processing_time also descreased.

Before	After

1 Raw Semseg Mask Image ~ 24000KB/s	Bandwidth: 1 Raw Semseg Mask Image ~ 320.72KB/s
https://github.com/user-attachments/assets/26b98e9c-83ff-46d3-95a3-e63246e4b53e	Screencast from 2024年07月26日 19時37分19秒.webm

Notes for reviewers

None.

Interface changes

None.

Effects on system behavior

None.

github-actions · 2024-07-09T00:29:34Z

Thank you for contributing to the Autoware project!

🚧 If your pull request is in progress, switch it to draft mode.

Please ensure:

You've checked our contribution guidelines.
Your PR follows our pull request guidelines.
All required CI checks pass before marking the PR ready for review.

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

codecov · 2024-07-26T08:12:04Z

Codecov Report

Attention: Patch coverage is 0% with 27 lines in your changes missing coverage. Please review.

Project coverage is 29.23%. Comparing base (3176fd7) to head (f7facd5).
Report is 18 commits behind head on main.

Files	Patch %	Lines
perception/autoware_tensorrt_yolox/src/utils.cpp	0.00%	22 Missing ⚠️
...utoware_tensorrt_yolox/src/tensorrt_yolox_node.cpp	0.00%	5 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #7905      +/-   ##
==========================================
- Coverage   29.25%   29.23%   -0.02%     
==========================================
  Files        1600     1601       +1     
  Lines      117736   117774      +38     
  Branches    50710    50729      +19     
==========================================
- Hits        34438    34430       -8     
- Misses      74123    74152      +29     
- Partials     9175     9192      +17

Flag	Coverage Δ		*Carryforward flag
differential	`0.00% <0.00%> (?)`
total	`29.24% <ø> (-0.01%)`	⬇️	Carriedforward from a853f8f

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

perception/autoware_tensorrt_yolox/src/utils.cpp

yukkysaito · 2024-07-29T01:29:47Z

@badai-nguyen Thanks for the PR.
Did you come up with the encoder-decoder being implemented this time by yourself?
I think PNG and JPEG (lossless mode) can be used since they are lossless compression formats, but I would like to know the reason if you didn't consider them this time.

badai-nguyen · 2024-07-29T08:15:11Z

@badai-nguyen Thanks for the PR. Did you come up with the encoder-decoder being implemented this time by yourself? I think PNG and JPEG (lossless mode) can be used since they are lossless compression formats, but I would like to know the reason if you didn't consider them this time.

@yukkysaito

The JPEG is not lossless compression so that can cause the change of segmentation mask, especiall along boundary of each segment.
I compared run_length with PNG from 'image_transport' and found that fatest compress level (level 1) of PNG still increase processing_time, and the 'run_length' shows the best performance.
Please check the updated PR's description.

Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com>

manato

@badai-nguyen
Thank you for your great contribution and sorry for the delay in my review. My major concern was increasing processing time by introducing run-length encoding. But thanks to the evaluation result you shared, it has gone.
I left two comments for small points. I'd appreciate it if you could consider them before approving this PR.

perception/autoware_tensorrt_yolox/src/utils.cpp

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com>

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

manato · 2024-08-07T07:43:07Z

perception/autoware_tensorrt_yolox/src/tensorrt_yolox_node.cpp

+    for (size_t i = 0; i < compressed_data.size(); ++i) {
+      std::memcpy(&out_mask_msg->data[i * step], &compressed_data.at(i).first, sizeof(uint8_t));
+      std::memcpy(
+        &out_mask_msg->data[i * step + sizeof(uint8_t)], &compressed_data.at(i).second,


same as #7905 (comment)

@manato I fixed at 3220095

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

manato

@badai-nguyen
LGTM. Thank you for handling all of my concerns. Since all my concerns are gone, I approve this PR.

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

badai-nguyen · 2024-08-08T13:21:43Z

Since seperated the unit test part into #8414

… mask (autowarefoundation#7905) * fix: add rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: move rle into utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: pre-commit Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> * fix: remove unused variable Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * style(pre-commit): autofix * feat: add unit test for utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: unit test Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: change to explicit index Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: cuda cmake Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: separate unit test into different PR Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

… mask (autowarefoundation#7905) * fix: add rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: move rle into utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: pre-commit Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> * fix: remove unused variable Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * style(pre-commit): autofix * feat: add unit test for utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: unit test Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: change to explicit index Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: cuda cmake Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: separate unit test into different PR Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: xtk8532704 <1041084556@qq.com>

… mask (autowarefoundation#7905) * fix: add rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: move rle into utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: pre-commit Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> * fix: remove unused variable Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * style(pre-commit): autofix * feat: add unit test for utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: unit test Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: change to explicit index Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: cuda cmake Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: separate unit test into different PR Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

….29.0 (#1526) * feat(tensorrt yolox): inference and publish mask image from yolox model with semantic segmentation header (autowarefoundation#5553) * add segmentation model Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> fix: add multitask for segment Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: publish mask Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * feat: publish colorized mask Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: resize yolox mask Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: add memory allocate operations Signed-off-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> * refactor: remove underscore for a local variable Signed-off-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> * chore: add condition to check the number of subscriber for newly added topics Signed-off-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> * chore: pre-commit Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: add roi overlapping segment Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: roi overlay segment Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: refactor Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * docs: update readme Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: update model name Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: add utils into tensorrt_yolox Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: launch file Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: remove unnecessary depend Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: fix yaml file Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: remove duplicated param in launch Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: semantic class Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * docs: update readme Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: update default param Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: add processing time topic Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: typo Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * chore: fix cspell error Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: yolox default param Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: rename debug topics Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: rename debug topics Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * docs: update model description Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: launch Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * refactor: unpublish mask for single task Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/tensorrt_yolox/src/tensorrt_yolox.cpp Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * Update perception/tensorrt_yolox/src/tensorrt_yolox.cpp Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * Update perception/tensorrt_yolox/src/tensorrt_yolox.cpp Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * style(pre-commit): autofix * docs: update reamde Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: skip mask size as yolox output Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Signed-off-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> Co-authored-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * fix(tensorrt_yolox): add run length encoding for sematic segmentation mask (autowarefoundation#7905) * fix: add rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: rle compress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: move rle into utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: pre-commit Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> * fix: remove unused variable Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_tensorrt_yolox/src/utils.cpp Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> * style(pre-commit): autofix * feat: add unit test for utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: unit test Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: change to explicit index Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: cuda cmake Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: separate unit test into different PR Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * refactor(tensorrt_yolox): move utils into perception_utils (autowarefoundation#8435) * chore(tensorrt_yolo): refactor utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: tensorrt_yolox Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * fix(perception_utils): install library after build (autowarefoundation#8501) Signed-off-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> * fix(image_projection_based_fusion): resize sematic segmentation mask as input image size (autowarefoundation#7635) Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix(image_projection_based_fusion): segmentation pointcloud fusion param update (autowarefoundation#7858) * fix(image_projection_based_fusion): add run length decoding for segmentation_pointcloud_fusion (autowarefoundation#7909) * fix: add rle decompress Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix: use rld in tensorrt utils Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: rebase error Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: dependency Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: skip publish debug mask Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Update perception/autoware_image_projection_based_fusion/src/segmentation_pointcloud_fusion/node.cpp Co-authored-by: kminoda <44218668+kminoda@users.noreply.github.com> * style(pre-commit): autofix * Revert "fix: skip publish debug mask" This reverts commit 30fa9ae. --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: kminoda <44218668+kminoda@users.noreply.github.com> * feat(tier4_perception_launch): add image segmentation based pointcloud filter (autowarefoundation#7225) * feat(tier4_perception_launch): add image segmentation based pointcloud filter Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: typo Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * fix: detection launch Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * chore: add maintainer Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * Revert "chore: add maintainer" This reverts commit 5adfef6. --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> * style(pre-commit): autofix * fix(autoware_image_projection_based_fusion): resolve issue with segmentation pointcloud fusion node failing with multiple mask inputs (autowarefoundation#8769) --------- Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp> Signed-off-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> Co-authored-by: Manato HIRABAYASHI <manato.hirabayashi@tier4.jp> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com> Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com> Co-authored-by: kminoda <44218668+kminoda@users.noreply.github.com> Co-authored-by: Yi-Hsiang Fang (Vivid) <146902905+vividf@users.noreply.github.com>

github-actions bot added the component:perception Advanced sensor data processing and environment understanding. (auto-assigned) label Jul 9, 2024

badai-nguyen mentioned this pull request Jul 9, 2024

fix(image_projection_based_fusion): add run length decoding for segmentation_pointcloud_fusion #7909

Merged

badai-nguyen changed the title ~~fix(tensorrt_yolox): add run length compressed for sematic segmentation mask~~ fix(tensorrt_yolox): add run length encoding for sematic segmentation mask Jul 10, 2024

badai-nguyen added 2 commits July 26, 2024 15:56

fix: add rle compress

f42cb41

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

fix: rle compress

10bfceb

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

badai-nguyen force-pushed the fix/add_run_length_compressed branch from 3d42d8b to 5f309f8 Compare July 26, 2024 07:32

fix: move rle into utils

c22d052

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

badai-nguyen force-pushed the fix/add_run_length_compressed branch from 6764334 to c22d052 Compare July 26, 2024 07:35

badai-nguyen added the run:build-and-test-differential Mark to enable build-and-test-differential workflow. (used-by-ci) label Jul 26, 2024

chore: pre-commit

a853f8f

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

badai-nguyen requested review from YoshiRi and yukkysaito July 26, 2024 11:05

badai-nguyen marked this pull request as ready for review July 26, 2024 11:05

badai-nguyen requested review from dan-dnn and manato as code owners July 26, 2024 11:05

yukkysaito reviewed Jul 29, 2024

View reviewed changes

perception/autoware_tensorrt_yolox/src/utils.cpp Outdated Show resolved Hide resolved

Update perception/autoware_tensorrt_yolox/src/utils.cpp

f7facd5

Co-authored-by: Yukihiro Saito <yukky.saito@gmail.com>

manato reviewed Aug 6, 2024

View reviewed changes

perception/autoware_tensorrt_yolox/src/utils.cpp Outdated Show resolved Hide resolved

perception/autoware_tensorrt_yolox/src/utils.cpp Outdated Show resolved Hide resolved

badai-nguyen and others added 5 commits August 7, 2024 09:19

fix: remove unused variable

12307ea

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

Update perception/autoware_tensorrt_yolox/src/utils.cpp

bd3f0b0

Co-authored-by: Manato Hirabayashi <3022416+manato@users.noreply.github.com>

style(pre-commit): autofix

e581b4d

feat: add unit test for utils

ce85d8b

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

style(pre-commit): autofix

1dc4020

badai-nguyen added run:build-and-test-differential Mark to enable build-and-test-differential workflow. (used-by-ci) and removed run:build-and-test-differential Mark to enable build-and-test-differential workflow. (used-by-ci) labels Aug 7, 2024

Merge branch 'main' into fix/add_run_length_compressed

7f8d44a

fix: unit test

44c6175

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

badai-nguyen force-pushed the fix/add_run_length_compressed branch from 2d837bf to 44c6175 Compare August 7, 2024 04:07

manato reviewed Aug 7, 2024

View reviewed changes

badai-nguyen and others added 4 commits August 7, 2024 16:54

chore: change to explicit index

3220095

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

style(pre-commit): autofix

191e2e3

fix: cuda cmake

a69a75a

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

Merge branch 'main' into fix/add_run_length_compressed

6c190e3

manato approved these changes Aug 8, 2024

View reviewed changes

badai-nguyen and others added 2 commits August 8, 2024 21:13

fix: separate unit test into different PR

1293d0d

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

Merge branch 'main' into fix/add_run_length_compressed

4ced050

badai-nguyen enabled auto-merge (squash) August 8, 2024 12:15

badai-nguyen merged commit dcd56fe into autowarefoundation:main Aug 8, 2024
22 of 24 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(tensorrt_yolox): add run length encoding for sematic segmentation mask #7905

fix(tensorrt_yolox): add run length encoding for sematic segmentation mask #7905

badai-nguyen commented Jul 9, 2024 •

edited

Loading

github-actions bot commented Jul 9, 2024 •

edited

Loading

codecov bot commented Jul 26, 2024 •

edited

Loading

yukkysaito commented Jul 29, 2024

badai-nguyen commented Jul 29, 2024

manato left a comment

manato Aug 7, 2024

badai-nguyen Aug 7, 2024

manato left a comment

badai-nguyen commented Aug 8, 2024

fix(tensorrt_yolox): add run length encoding for sematic segmentation mask #7905

fix(tensorrt_yolox): add run length encoding for sematic segmentation mask #7905

Conversation

badai-nguyen commented Jul 9, 2024 • edited Loading

Description

Related links

How was this PR tested?

Notes for reviewers

Interface changes

Effects on system behavior

github-actions bot commented Jul 9, 2024 • edited Loading

codecov bot commented Jul 26, 2024 • edited Loading

Codecov Report

yukkysaito commented Jul 29, 2024

badai-nguyen commented Jul 29, 2024

manato left a comment

Choose a reason for hiding this comment

manato Aug 7, 2024

Choose a reason for hiding this comment

badai-nguyen Aug 7, 2024

Choose a reason for hiding this comment

manato left a comment

Choose a reason for hiding this comment

badai-nguyen commented Aug 8, 2024

badai-nguyen commented Jul 9, 2024 •

edited

Loading

github-actions bot commented Jul 9, 2024 •

edited

Loading

codecov bot commented Jul 26, 2024 •

edited

Loading