Update readme and fix data names

README.md

@@ -36,7 +36,8 @@ pip install nuscenes-devkit openpifpaf
     Data         
     ├── arrays                 
     ├── models
-    ├── baselines
+    ├── kitti
+    ├── nuscenes
     ├── logs
     
 
@@ -44,7 +45,7 @@ Run the following to create the folders:
 ```
 mkdir data
 cd data
-mkdir arrays models baselines logs
+mkdir arrays models data-kitti data-nuscenes logs
 ```
 
 ### Pre-trained Models
@@ -52,11 +53,6 @@ mkdir arrays models baselines logs
 * Download a Pifpaf pre-trained model from [openpifpaf](https://github.com/vita-epfl/openpifpaf) project 
 and save it into `data/models`
 
-### Baselines
-Download KITTI ground truth txt files from Google Drive ADD link and unzip them in `data/baselines`
-The zip file also contains nuScenes and KITTI splits for training and validations as well as detections
-from other baselines (more details in Eval section)
-
 
 # Interfaces
 All the commands are run through a main file called `main.py` using subparsers.
@@ -102,6 +98,22 @@ Without ground_truth matching (all the detected people):
 ![predict_no_matching](docs/002282.png.combined_2.png)
 
 # Preprocess
+
+### Datasets
+
+#### 1) KITTI dataset
+Download KITTI ground truth files and camera calibration matrices for training
+from [here](http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d) and
+save them respectively into `data/kitti/gt` and `data/kitti/calib`. 
+To extract pifpaf joints, you also need to download training images, put it in any folder and soft link in `
+data/kitti/images`
+
+#### 2) nuScenes dataset
+Download nuScenes dataset (any version: Mini, Teaser or TrainVal) from [nuScenes](https://www.nuscenes.org/download), 
+save it anywhere and soft link it in `data/nuscenes`
+
+
+
 ### Input joints for training
 MonoLoco is trained using 2D human pose joints detected by pifpaf and matched with the ground truth location provided by
 nuScenes or KITTI Dataset. To create the joints run: `python src/main.py prep` specifying:
@@ -112,7 +124,12 @@ dataset are supported: nuscenes_mini, nuscenes, nuscenes_teaser.
 
 ### Ground truth file for evaluation
 The preprocessing script also outputs a second json file called **names.json** which provide a dictionary indexed
-by the image name to easily access ground truth files for evaluation purposes.
+by the image name to easily access ground truth files for evaluation and prediction purposes.
+
+
+
+
+
 
 
 ### Train
@@ -134,6 +151,9 @@ and stereo Baselines:
 [MonoDepth](https://arxiv.org/abs/1609.03677) and our 
 [Geometrical Baseline](src/eval/geom_baseline.py).
 
+Alternatively we provide the links to download them.
+
+
 The following graph is obtained running:
 `python3 src/main.py eval --dataset kitti --model data/models/base_model.pickle`
 ![kitti_evaluation](docs/results.png)

src/eval/kitti_eval.py

@@ -30,13 +30,13 @@ class KittiEval:
         from visuals.results import print_results
         self.print_results = print_results
 
-        self.dir_gt = os.path.join('data', 'baselines', 'gt')
+        self.dir_gt = os.path.join('data', 'kitti', 'gt')
-        self.dir_m3d = os.path.join('data', 'baselines', 'm3d')
+        self.dir_m3d = os.path.join('data', 'kitti', 'm3d')
-        self.dir_3dop = os.path.join('data', 'baselines', '3dop')
+        self.dir_3dop = os.path.join('data', 'kitti', '3dop')
-        self.dir_md = os.path.join('data', 'baselines', 'monodepth')
+        self.dir_md = os.path.join('data', 'kitti', 'monodepth')
-        self.dir_psm = os.path.join('data', 'baselines', 'psm')
+        self.dir_psm = os.path.join('data', 'kitti', 'psm')
-        self.dir_our = os.path.join('data', 'baselines', 'monoloco')
+        self.dir_our = os.path.join('data', 'kitti', 'monoloco')
-        path_val = os.path.join('data', 'baselines', 'val.txt')
+        path_val = os.path.join('data', 'kitti', 'val.txt')
         dir_logs = os.path.join('data', 'logs')
         assert dir_logs, "No directory to save final statistics"
 

src/features/preprocess_ki.py

@@ -15,7 +15,7 @@ class PreprocessKitti:
 
         self.dir_ann = dir_ann
         self.iou_thresh = iou_thresh
-        self.dir_gt = os.path.join('data', 'baselines', 'gt')
+        self.dir_gt = os.path.join('data', 'kitti', 'gt')
         self.names_gt = os.listdir(self.dir_gt)
         self.dir_kk = os.path.join('data', 'baselines', 'calib')
         self.list_gt = glob.glob(self.dir_gt + '/*.txt')

src/main.py

@@ -10,7 +10,6 @@ from openpifpaf import decoder
 from features.preprocess_nu import PreprocessNuscenes
 from features.preprocess_ki import PreprocessKitti
 from predict.predict_2d_3d import predict
-from features.trial import trials
 from models.trainer import Trainer
 from eval.run_kitti import RunKitti
 from eval.geom_baseline import GeomBaseline
@@ -62,7 +61,7 @@ def cli():
                                 default="data/models/best_model__seed_2_.pickle")
     predict_parser.add_argument('--path_gt', help='path of json file with gt 3d localization',
                                 default='data/arrays/names-kitti-000.json')
-    predict_parser.add_argument('--dir_calib', default='data/baselines/calib/', help='directory of calib_files')
+    predict_parser.add_argument('--dir_calib', default='data/kitti/calib/', help='directory of calib_files')
     predict_parser.add_argument('--transform', help='transformation for the pose', default='None')
     predict_parser.add_argument('--draw_kps', help='to draw kps in the images', action='store_true')
     predict_parser.add_argument('--predict', help='whether to make prediction', action='store_true')