charles/monoloco
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

update for experiments and readme

Browse Source
This commit is contained in:
Lorenzo 2020-12-07 15:28:54 +01:00
parent 810624a976
commit bf727c03c8
7 changed files with 65 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
LICENSE
View File

@ -1,4 +1,4 @@
Copyright 2020 by EPFL/VITA. All rights reserved.
Copyright 2020-2021 by EPFL/VITA. All rights reserved.
This project and all its files are licensed under
GNU AGPLv3 or later version.
@ -6,4 +6,4 @@ GNU AGPLv3 or later version.
If this license is not suitable for your business or project
please contact EPFL-TTO (https://tto.epfl.ch/) for a full commercial license.
This software may not be used to harm any person deliberately.
This software may not be used to harm any person deliberately or for any military application.

51
docs/MonoLoco++.md
View File

@ -82,4 +82,53 @@ Our results are obtained with:
For a more extensive list of available parameters, run:
`python -m monstereo.run train --help`
`python -m monstereo.run train --help`
## Evaluation
### 3D Localization
We provide evaluation on KITTI for models trained on nuScenes or KITTI. We compare them with other monocular
and stereo Baselines:
[MonoLoco](TODO),
[Mono3D](https://www.cs.toronto.edu/~urtasun/publications/chen_etal_cvpr16.pdf),
[3DOP](https://xiaozhichen.github.io/papers/nips15chen.pdf),
[MonoDepth](https://arxiv.org/abs/1609.03677)
[MonoPSR](TODO) and our
[MonoDIS](TODO) and our
[Geometrical Baseline](monoloco/eval/geom_baseline.py).
* **Mono3D**: download validation files from [here](http://3dimage.ee.tsinghua.edu.cn/cxz/mono3d)
and save them into `data/kitti/m3d`
* **3DOP**: download validation files from [here](https://xiaozhichen.github.io/)
and save them into `data/kitti/3dop`
* **MonoDepth**: compute an average depth for every instance using the following script
[here](https://github.com/Parrotlife/pedestrianDepth-baseline/tree/master/MonoDepth-PyTorch)
and save them into `data/kitti/monodepth`
* **GeometricalBaseline**: A geometrical baseline comparison is provided.
Download the model for monoloco
The average geometrical value for comparison can be obtained running:
```
python -m monstereo.run eval
--dir_ann <annotation directory>
--model <model path>
--net monoloco_pp
--generate
````
To include also geometric baselines and MonoLoco, add the flag ``--baselines``
### 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)
```
python -m monstereo.run eval
--activity
--dataset collective
--net monoloco_pp
--model <path to the model>
--dir_ann <annotation directory>
```

10
monstereo/activity.py
View File

@ -16,7 +16,7 @@ from matplotlib.patches import Circle, FancyArrow
from PIL import Image
from .network.process import laplace_sampling
from .utils import open_annotations
from .utils import open_annotations, get_task_error
from .visuals.pifpaf_show import KeypointPainter, image_canvas
from .network import Loco
from .network.process import factory_for_gt, preprocess_pifpaf
@ -37,7 +37,7 @@ def social_interactions(idx, centers, angles, dds, stds=None, social_distance=Fa
if n_samples < 2:
for idx_t in indices:
if check_f_formations(idx, idx_t, centers, angles,
radii=radii, # Binary value
radii=radii, # Binary value
social_distance=social_distance):
return True
@ -46,7 +46,7 @@ def social_interactions(idx, centers, angles, dds, stds=None, social_distance=Fa
# Samples distance
dds = torch.tensor(dds).view(-1, 1)
stds = torch.tensor(stds).view(-1, 1)
# stds_te = get_task_error(dds) # similar results to MonoLoco but lower true positive
# stds = get_task_error(dds) # similar results to MonoLoco but lower true positive
laplace_d = torch.cat((dds, stds), dim=1)
samples_d = laplace_sampling(laplace_d, n_samples=n_samples)
@ -72,8 +72,8 @@ def social_interactions(idx, centers, angles, dds, stds=None, social_distance=Fa
def check_f_formations(idx, idx_t, centers, angles, radii, social_distance=False):
"""
Check F-formations for people close together:
1) Empty space of 0.4 + meters (no other people or themselves inside)
Check F-formations for people close together (this function do not expect far away people):
1) Empty space of a certain radius (no other people or themselves inside)
2) People looking inward
"""

6
monstereo/eval/eval_activity.py
View File

@ -32,10 +32,10 @@ class ActivityEvaluator:
self.sequences = ['seq02', 'seq14', 'seq12', 'seq13', 'seq11', 'seq36']
# folders_collective = ['seq02']
self.dir_data = 'data/activity/dataset'
self.THRESHOLD_PROB = 0.25 # Concordance for samples
self.THRESHOLD_PROB = 0.2 # Concordance for samples
self.THRESHOLD_DIST = 2 # Threshold to check distance of people
self.RADII = (0.3, 0.5) # expected radii of the o-space
self.PIFPAF_CONF = 0.4
self.PIFPAF_CONF = 0.3
self.SOCIAL_DISTANCE = False
# -------------------------------------------------------------------------------------------------------------
@ -76,7 +76,7 @@ class ActivityEvaluator:
# Collect PifPaf files and calibration
basename = os.path.basename(im_path)
extension = '.pifpaf.json'
extension = '.predictions.json'
path_pif = os.path.join(self.dir_ann, basename + extension)
annotations = open_annotations(path_pif)
kk, _ = factory_for_gt(im_size, verbose=False)

1
monstereo/eval/generate_kitti.py
View File

@ -241,6 +241,7 @@ def save_txts(path_txt, all_inputs, all_outputs, all_params, mode='monoloco', ca
conf_scale = 0.03
elif mode == 'monoloco_pp':
conf_scale = 0.033
# conf_scale = 0.036 nuScenes for having same recall
else:
conf_scale = 0.055
conf = conf_scale * (uv_box[-1]) / (bi / math.sqrt(xx ** 2 + yy ** 2 + zz ** 2))

8
monstereo/prep/preprocess_nu.py
View File

@ -87,7 +87,7 @@ class PreprocessNuscenes:
while not current_token == "":
sample_dic = self.nusc.get('sample', current_token)
cnt_samples += 1
# if (cnt_samples % 4 == 0) and (cnt_ann < 3000):
# Extract all the sample_data tokens for each sample
for cam in self.CAMERAS:
sd_token = sample_dic['data'][cam]
@ -105,7 +105,7 @@ class PreprocessNuscenes:
self.dic_names[basename + '.jpg']['K'] = copy.deepcopy(kk)
# Run IoU with pifpaf detections and save
path_pif = os.path.join(self.dir_ann, name + '.pifpaf.json')
path_pif = os.path.join(self.dir_ann, name + '.predictions.json')
exists = os.path.isfile(path_pif)
if exists:
@ -114,7 +114,6 @@ class PreprocessNuscenes:
boxes, keypoints = preprocess_pifpaf(annotations, im_size=(1600, 900))
else:
continue
if keypoints:
matches = get_iou_matches(boxes, boxes_gt, self.iou_min)
for (idx, idx_gt) in matches:
@ -130,7 +129,6 @@ class PreprocessNuscenes:
append_cluster(self.dic_jo, phase, inp, lab, keypoint)
cnt_ann += 1
sys.stdout.write('\r' + 'Saved annotations {}'.format(cnt_ann) + '\t')
current_token = sample_dic['next']
with open(os.path.join(self.path_joints), 'w') as f:
@ -139,7 +137,7 @@ class PreprocessNuscenes:
json.dump(self.dic_names, f)
end = time.time()
extract_box_average(self.dic_jo['train']['boxes_3d'])
# extract_box_average(self.dic_jo['train']['boxes_3d'])
print("\nSaved {} annotations for {} samples in {} scenes. Total time: {:.1f} minutes"
.format(cnt_ann, cnt_samples, cnt_scenes, (end-start)/60))
print("\nOutput files:\n{}\n{}\n".format(self.path_names, self.path_joints))

4
setup.py
View File

@ -27,9 +27,7 @@ setup(
zip_safe=False,
install_requires=[
'openpifpaf==0.8.0',
'torch==1.1.0',
'torchvision==0.3.0'
'openpifpaf>=0.11'
],
extras_require={
'eval': [