Skip to content

Compare ground truth investigation (#251)#253

Closed
chanwutk wants to merge 2 commits intooptimized_ingestionfrom
compare-ground-truth-investigation-2
Closed

Compare ground truth investigation (#251)#253
chanwutk wants to merge 2 commits intooptimized_ingestionfrom
compare-ground-truth-investigation-2

Conversation

@chanwutk
Copy link
Copy Markdown
Contributor

@chanwutk chanwutk commented Aug 6, 2023

No description provided.

chanwutk and others added 2 commits August 6, 2023 16:42
* add helper functions for detection algorithm

* style: [CI] format

* algorithms necessary functions finish, integration testing not completely done

* delete trailing lines

* add test

* finish all implementation of sample plan, unit tested through notebook

* end to end flow succeed, integration testing ongoing

* thorough run of videos

* fix ego exit segment

* fix small bug

* fix detection problem after ego exit fix

* style: [CI] format

* add doc string

* revert data files

* .

* typing

* annotate typing

* add necessary typing

* style: [CI] format

* add ego stop action

* fix import

* Integrate detection estimation to pipeline (#186)

* .

* style: [CI] format

* split yolo and strongsort

* add detection estimation

* style: [CI] format

* style: [CI] format

* run detection estimation with pipeline

* style: [CI] format

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>

* fix errors

* fix error

* clean up

* style: [CI] format

* detection estimation as stage

* style: [CI] format

* add detection_estimation to notebook

* style: [CI] format

* use logger instead of print

* style: [CI] format

* fix typing in python3.8

* style: [CI] format

* clean up

* style: [CI] format

* add TODO for detection coversio

* Add 3d calibration against the road segment (#187)

* add to do of car_loc3d conversion

* tested, including segment line in the road segment

* add location calibration

* style: [CI] format

* fix flake errors

* Update __init__.py

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>
Co-authored-by: Chanwut (Mick) Kittivorawong <30903997+chanwutk@users.noreply.github.com>

* benchmark

* fix LineString error + output trajectories for benchmark

* style: [CI] format

* performance benchmark

* Detection Estimation: add 3D detection and output DetectionInfo (#189)

* add 3d detection and DetectionInfo

* style: [CI] format

* fix numpy operations

* update nb

* style: [CI] format

* document 3d detection dimensions

* style: [CI] format

* fix lint error

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>

* fix algorithms with new discovery in segment lines (#188)

* fix algorithms with new discovery in segment lines

* style: [CI] format

* include intersection in detection info so that it could be passed down

* style: [CI] format

* fix lint error

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>
Co-authored-by: Chanwut (Mick) Kittivorawong <30903997+chanwutk@users.noreply.github.com>
Co-authored-by: Chanwut (Mick) Kittivorawong <chanwutk@gmail.com>

* cleanup duplicate road segments

* style: [CI] format

* allow none uid

* fix test

* fix linting error

* video with skipped frames

* style: [CI] format

* Construct segment trajectory (#191)

* add to do of car_loc3d conversion

* tested, including segment line in the road segment

* add location calibration

* style: [CI] format

* fix bug in intersection extended lines

* add stage of constructing segment trajectory

* style: [CI] format

* add docstring

* fix docstring

* style: [CI] format

* fix function naming

* style: [CI] format

* fix import

* style: [CI] format

* typing

* style: [CI] format

* add complete story algorithm

* style: [CI] format

* initial push of the segment trajectory stage

* style: [CI] format

* integrate segment trajectory to pipeline (#192)

* integrate to pipeline

* style: [CI] format

* fix type

* run benchmark

* style: [CI] format

* clean up and update instructions

* clean up playground

* style: [CI] cleanup python notebooks

* update dir

* style: [CI] cleanup python notebooks

* fix errors

* style: [CI] cleanup python notebooks

* style: [CI] format

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>

* fix lint error

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>
Co-authored-by: Chanwut (Mick) Kittivorawong <chanwutk@gmail.com>
Co-authored-by: Chanwut (Mick) Kittivorawong <30903997+chanwutk@users.noreply.github.com>

* add detection type annotation

* fix detection estimation errors (#195)

* fix detection estimation errors

* style: [CI] cleanup python notebooks

* style: [CI] format

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>

* Add object type filter (#196)

* add type filter

* style: [CI] format

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>

* Fix Detection Estimation segment mapping (#197)

* fix polygon contains

* style: [CI] format

* fix postgis LineString[] typing

* style: [CI] cleanup python notebooks

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>

* pruning testing of detection estimation (#199)

* pruning testing of detection estimation

* style: [CI] format

* add type

* style: [CI] format

Co-authored-by: Apperception Service <apperception@freddie.Millennium.Berkeley.edu>
Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>
Co-authored-by: Chanwut (Mick) Kittivorawong <chanwutk@gmail.com>

* fix the pipeline notebook

* benchmark on front cameras

* Optimized mapping (#200)

* new mapping

* style: [CI] format

* disable pruning and fix pipeline

* fix detection estimation pipeline for new mapping

* style: [CI] cleanup python notebooks

* new mapping benchmark

* clean up comments

* style: [CI] cleanup python notebooks

* style: [CI] format

* enable cache in strongsort

* style: [CI] format

* add customized predicate for detection_estimation to prune detection, and add fig14 pipeline

* style: [CI] cleanup python notebooks

* style: [CI] format

---------

Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>
Co-authored-by: Apperception Service <apperception@freddie.Millennium.Berkeley.EDU>

* Fix road type (#201)

* fix the pipeline notebook

* benchmark on front cameras

* fix road type

* style: [CI] cleanup python notebooks

---------

Co-authored-by: Apperception Service <apperception@freddie.Millennium.Berkeley.EDU>
Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>

* finish final merge

* add end to end fig 14 and fig 15, along with results of fig13 compared with groundtruth

* add helper files for investigation

* style: [CI] cleanup python notebooks

* style: [CI] format

* add support to run full mini

* fixed fig15

* style: [CI] cleanup python notebooks

* style: [CI] format

* added notebook to visualize images + fixed figures 13 and 14

* style: [CI] cleanup python notebooks

* updated visualize image notebook

* full dataset applied

* style: [CI] cleanup python notebooks

* timing added

* some more devkit fixes

* nuscenes comparison done

* reupdate

* published

* style: [CI] format

---------

Co-authored-by: YongmingFrankGe <frankge_1012@outlook.com>
Co-authored-by: Github Actions Bot <apperception-actions-bot@users.noreply.github.com>
Co-authored-by: Yongming Ge <46760841+YongmingFrankGe@users.noreply.github.com>
Co-authored-by: Apperception Service <apperception@freddie.Millennium.Berkeley.edu>
Co-authored-by: Yousef Helal <yousefh@berkeley.edu>
@codecov
Copy link
Copy Markdown

codecov bot commented Aug 6, 2023

Codecov Report

Patch and project coverage have no change.

Comparison is base (2673817) 64.33% compared to head (89e79c8) 64.33%.

Additional details and impacted files
@@                 Coverage Diff                  @@
##           optimized_ingestion     #253   +/-   ##
====================================================
  Coverage                64.33%   64.33%           
====================================================
  Files                       63       63           
  Lines                     2013     2013           
====================================================
  Hits                      1295     1295           
  Misses                     718      718           

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

@chanwutk
Copy link
Copy Markdown
Contributor Author

chanwutk commented Aug 7, 2023

merged in spatialyze

@chanwutk chanwutk closed this Aug 7, 2023
@chanwutk chanwutk deleted the compare-ground-truth-investigation-2 branch August 7, 2023 00:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant