diff --git a/README.md b/README.md index d1b59c4..1a2fcb1 100644 --- a/README.md +++ b/README.md @@ -189,10 +189,7 @@ docs/frame0038.jpg \ --output_types front bird --show_all \ --model data/models/monoloco_pp-201203-1424.pkl -o ``` - - - - + @@ -354,35 +351,35 @@ and save them into `data/kitti/monodepth` * **Geometrical Baseline and MonoLoco**: To include also geometric baselines and MonoLoco, add the flag ``--baselines`` to the evaluation command ``` -python -m monstereo.run eval ---dir_ann ---model ---net monoloco_pp ---generate +python -m monoloco.run eval \ +--dir_ann \ +--model \ +--net monoloco_pp \ +--generate \ +--save \ ```` -### Relative Average Precision Localization (RALP-5%) +#TODO add white table for monocular + + +By changing the net and the model, the same command evaluates MonStereo model. + +#TODO Add white table for stereo + + + +### Relative Average Precision Localization: RALP-5% (MonStereo) We modified the original C++ evaluation of KITTI to make it relative to distance. We use **cmake**. -To run the evaluation, first generate the txt files with: - -`python -m monstereo.run eval --dir_ann --model data/models/ms-200710-1511.pkl --generate` - +To run the evaluation, first generate the txt file with the standard command for evaluation (above). Then follow the instructions of this [repository](https://github.com/cguindel/eval_kitti) to prepare the folders accordingly (or follow kitti guidelines) and run evaluation. The modified file is called *evaluate_object.cpp* and runs exactly as the original kitti evaluation. - - - - -Adding the argument `save`, a few plots will be added including 3D localization error as a function of distance: - - ### Activity Estimation (Talking) Please follow preprocessing steps for Collective activity dataset and run pifpaf over the dataset images. Evaluation on this dataset is done with models trained on either KITTI or nuScenes. -For optimal performances, we suggest the model trained on nuScenes teaser (TODO add link) +For optimal performances, we suggest the model trained on nuScenes teaser (#TODO add link) ``` python -m monstereo.run eval --activity @@ -393,24 +390,6 @@ python -m monstereo.run eval ``` - -### Data structure - - Data - ├── arrays - ├── models - ├── kitti - ├── figures - ├── logs - - -Run the following to create the folders: -``` -mkdir data -cd data -mkdir arrays models kitti figures logs -``` - ## Citation When using this library in your research, we will be happy if you cite us! @@ -423,7 +402,6 @@ When using this library in your research, we will be happy if you cite us! } ``` - ``` @ARTICLE{bertoni_2021_its, author = {Bertoni, Lorenzo and Kreiss, Sven and Alahi, Alexandre}, @@ -432,7 +410,6 @@ When using this library in your research, we will be happy if you cite us! year={2021}, ``` - ``` @InProceedings{bertoni_2019_iccv, author = {Bertoni, Lorenzo and Kreiss, Sven and Alahi, Alexandre}, diff --git a/docs/out_frame0038.jpg.bird.jpg b/docs/out_frame0038.jpg.bird.jpg deleted file mode 100644 index ab49eb1..0000000 Binary files a/docs/out_frame0038.jpg.bird.jpg and /dev/null differ diff --git a/docs/out_frame0038.jpg.front.jpg b/docs/out_frame0038.jpg.front.jpg deleted file mode 100644 index e01fcdf..0000000 Binary files a/docs/out_frame0038.jpg.front.jpg and /dev/null differ diff --git a/docs/out_frame0038.jpg.front_bird.jpg b/docs/out_frame0038.jpg.front_bird.jpg new file mode 100644 index 0000000..3fbfaac Binary files /dev/null and b/docs/out_frame0038.jpg.front_bird.jpg differ diff --git a/docs/results.png b/docs/results.png deleted file mode 100644 index 2a092c0..0000000 Binary files a/docs/results.png and /dev/null differ diff --git a/docs/results_monoloco_pp.jpg b/docs/results_monoloco_pp.jpg new file mode 100644 index 0000000..4fb40bd Binary files /dev/null and b/docs/results_monoloco_pp.jpg differ diff --git a/docs/results_stereo.jpg b/docs/results_stereo.jpg new file mode 100644 index 0000000..bddef0f Binary files /dev/null and b/docs/results_stereo.jpg differ