add adabins and yolo

Author: Stippler

.gitignore

@@ -128,4 +128,4 @@ dmypy.json
 # Pyre type checker
 .pyre/
 
-res/models
+res/*

.vscode/settings.json

@@ -0,0 +1,8 @@
+{
+    "python.linting.pylintArgs": [
+        "--generated-members=numpy.* ,torch.*"
+    ],
+    "python.linting.pylintEnabled": false,
+    "python.linting.flake8Enabled": true,
+    "python.linting.enabled": true
+}

__init__.py

@@ -1,9 +1,8 @@
-#%%
 from models.adabins.infer import InferenceHelper
-import matplotlib.pyplot as plt
 from time import time
 import cv2
 from PIL import Image
+import torch
 
 # img = Image.open("test_imgs/classroom__rgb_00283.jpg")
 start = time()
@@ -12,14 +11,12 @@
 cap = cv2.VideoCapture('https://192.168.0.164:8080/video')
 while True:
     ret, frame = cap.read()
-    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)/255
 
     im_pil = Image.fromarray(frame)
     # tgt_img = load_tensor_image( dataset_dir + test_files[j], args)
     centers, pred = inferHelper.predict_pil(im_pil)
-    depth_img = pred[0,0]
-    print(depth_img.shape)
-    print(1/depth_img)
+    depth_img = pred[0, 0]/10
     cv2.imshow('frame', depth_img)
     if cv2.waitKey(1) & 0xFF == ord('q'):
         break

adabins.py

@@ -8,9 +8,9 @@
 from torchvision import transforms
 from tqdm import tqdm
 
-import models.adabins.model_io as model_io
-import models.adabins.utils as utils
-from models.adabins.models import UnetAdaptiveBins
+import adabins.model_io as model_io
+import adabins.utils as utils
+from adabins.models import UnetAdaptiveBins
 
 
 def _is_pil_image(img):
@@ -72,7 +72,7 @@ def __init__(self, dataset='nyu', device='cuda:0'):
             self.max_depth = 10
             self.saving_factor = 1000  # used to save in 16 bit
             model = UnetAdaptiveBins.build(n_bins=256, min_val=self.min_depth, max_val=self.max_depth)
-            pretrained_path = "./res/models/adabins_nyu.pt"
+            pretrained_path = "./res/weights/adabins_nyu.pt"
         elif dataset == 'kitti':
             raise ValueError("dataset can be only 'nyu' {}".format(dataset))
             self.min_depth = 1e-3

models/adabins/__init__.py adabins/__init__.py

@@ -0,0 +1,60 @@
+from matplotlib import pyplot as plt
+from utils import predict, load_images, display_images
+from layers import BilinearUpSampling2D
+from keras.models import load_model
+import os
+import glob
+import argparse
+import matplotlib
+
+# Keras / TensorFlow
+os.environ['TF_CPP_MIN_LOG_LEVEL'] = '5'
+
+# Argument Parser
+parser = argparse.ArgumentParser(
+    description='High Quality Monocular Depth Estimation via Transfer Learning')
+parser.add_argument('--model', default='nyu.h5', type=str,
+                    help='Trained Keras model file.')
+parser.add_argument('--input', default='examples/*.png',
+                    type=str, help='Input filename or folder.')
+args = parser.parse_args()
+
+# Custom object needed for inference and training
+custom_objects = {'BilinearUpSampling2D': BilinearUpSampling2D,
+                  'depth_loss_function': None}
+
+print('Loading model...')
+
+# Load model into GPU / CPU
+model = load_model(args.model, custom_objects=custom_objects, compile=False)
+
+print('\nModel loaded ({0}).'.format(args.model))
+
+# Input images
+inputs = load_images(glob.glob(args.input))
+print('\nLoaded ({0}) images of size {1}.'.format(
+    inputs.shape[0], inputs.shape[1:]))
+
+# Compute results
+outputs = predict(model, inputs)
+
+# matplotlib problem on ubuntu terminal fix
+# matplotlib.use('TkAgg')
+
+# Display results
+viz = display_images(outputs.copy(), inputs.copy())
+plt.figure(figsize=(10, 5))
+plt.imshow(viz)
+plt.savefig('test.png')
+plt.show()
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--weights', nargs='+', type=str,
+                        default='res/weights/yolov3.pt', help='model.pt path(s)')
+    # file/folder, 0 for webcam
+    parser.add_argument('--source', type=str,
+                        default='data/images', help='source')
+    opt = parser.parse_args()
+    print(opt)