added requirements file

Author: RSKothari

README.md

@@ -44,7 +44,7 @@ Note the other flags and their (default values)
 * `--save_overlay` (True) - Save out segmentation output map on eye video
 * `--eval_on_cpu` (False) - If no GPU available or found, EllSeg can also be evaluated using CPU
 * `--load_file` (./weights/all.git_ok) - Choose a weight configuration. Default is *all* which was trained on a combination of all available datasets 
-* `--check_for_string_in_fname` ('') - Only evaluate on videos with a user defined string. Example `--check_for_string_in_fname=eye` will evaluate on eye videos only
+* `--check_for_string_in_fname` ('') - Only evaluate on videos with a user defined string. Example `--check_for_string_in_fname=eye` will evaluate on videos with <eye> within it
 
 # Try it out on your eye images!
 Coming soon!

evaluate_ellseg.py

@@ -209,6 +209,9 @@ def evaluate_ellseg_per_video(path_vid, args, model):
     fourcc = cv2.VideoWriter_fourcc(*'mp4v')
     vid_out = cv2.VideoWriter(path_vid_out, fourcc, int(FR), (int(W), int(H)))
 
+    # Dictionary to save output ellipses
+    ellipse_out_dict = {}
+
     ret = True
     pbar = tqdm(total=FR_COUNT)
 
@@ -226,19 +229,29 @@ def evaluate_ellseg_per_video(path_vid, args, model):
         seg_map, latent, pupil_ellipse, iris_ellipse = evaluate_ellseg_on_image(input_tensor, model)
 
         # Return ellipse predictions back to original dimensions
-        seg_map, pupil_ellipse, iris_ellipse = rescale_to_original(seg_map, pupil_ellipse, iris_ellipse, scale_shift, frame.shape)
+        seg_map, pupil_ellipse, iris_ellipse = rescale_to_original(seg_map,
+                                                                   pupil_ellipse,
+                                                                   iris_ellipse,
+                                                                   scale_shift,
+                                                                   frame.shape)
 
         # Generate visuals
         frame_overlayed_with_op = plot_segmap_ellpreds(frame, seg_map, pupil_ellipse, iris_ellipse)
         vid_out.write(frame_overlayed_with_op[..., ::-1])
 
+        # Append to dictionary
+        ellipse_out_dict[counter] = {'pupil': pupil_ellipse, 'iris': iris_ellipse}
+
         pbar.update(1)
         counter+=1
 
     vid_out.close()
     vid_obj.close()
     pbar.close()
 
+    # Save out ellipse dictionary
+    np.save(os.path.join(path_dir, file_name+'_pred.npy'), ellipse_out_dict)
+
     return True
 
 

requirements.txt

@@ -0,0 +1,108 @@
+# This file may be used to create an environment using:
+# $ conda create --name <env> --file <this file>
+# platform: linux-64
+_libgcc_mutex=0.1=main
+blas=1.0=mkl
+blosc=1.20.1=he1b5a44_0
+bzip2=1.0.8=h7b6447c_0
+ca-certificates=2021.1.19=h06a4308_0
+cairo=1.14.12=h8948797_3
+certifi=2020.12.5=py36h5fab9bb_1
+cffi=1.14.4=py36h261ae71_0
+cloudpickle=1.6.0=py_0
+cudatoolkit=9.2=0
+cycler=0.10.0=py36_0
+cytoolz=0.11.0=py36h7b6447c_0
+dask-core=2021.1.1=pyhd3eb1b0_0
+dbus=1.13.18=hb2f20db_0
+decorator=4.4.2=py_0
+deepdish=0.3.6=pyh9f0ad1d_0
+expat=2.2.10=he6710b0_2
+ffmpeg=4.0=hcdf2ecd_0
+fontconfig=2.13.0=h9420a91_0
+freeglut=3.0.0=hf484d3e_5
+freetype=2.10.4=h5ab3b9f_0
+glib=2.66.1=h92f7085_0
+graphite2=1.3.14=h23475e2_0
+gst-plugins-base=1.14.0=h8213a91_2
+gstreamer=1.14.0=h28cd5cc_2
+h5py=2.8.0=py36h989c5e5_3
+harfbuzz=1.8.8=hffaf4a1_0
+hdf5=1.10.2=hba1933b_1
+icu=58.2=he6710b0_3
+imageio=2.9.0=py_0
+intel-openmp=2020.2=254
+jasper=2.0.14=h07fcdf6_1
+joblib=1.0.0=pyhd3eb1b0_0
+jpeg=9b=h024ee3a_2
+kiwisolver=1.3.0=py36h2531618_0
+lcms2=2.11=h396b838_0
+ld_impl_linux-64=2.33.1=h53a641e_7
+libedit=3.1.20191231=h14c3975_1
+libffi=3.3=he6710b0_2
+libgcc-ng=9.1.0=hdf63c60_0
+libgfortran-ng=7.3.0=hdf63c60_0
+libglu=9.0.0=hf484d3e_1
+libopencv=3.4.2=hb342d67_1
+libopus=1.3.1=h7b6447c_0
+libpng=1.6.37=hbc83047_0
+libstdcxx-ng=9.1.0=hdf63c60_0
+libtiff=4.1.0=h2733197_1
+libuuid=1.0.3=h1bed415_2
+libvpx=1.7.0=h439df22_0
+libxcb=1.14=h7b6447c_0
+libxml2=2.9.10=hb55368b_3
+lz4-c=1.9.3=h2531618_0
+lzo=2.10=h516909a_1000
+matplotlib=3.1.0=py36h5429711_0
+matplotlib-base=3.3.2=py36h817c723_0
+mkl=2020.2=256
+mkl-service=2.3.0=py36he8ac12f_0
+mkl_fft=1.2.0=py36h23d657b_0
+mkl_random=1.1.1=py36h0573a6f_0
+ncurses=6.2=he6710b0_1
+networkx=2.5=py_0
+ninja=1.10.2=py36hff7bd54_0
+numexpr=2.7.1=py36h7c3b610_3
+numpy=1.19.2=py36h54aff64_0
+numpy-base=1.19.2=py36hfa32c7d_0
+olefile=0.46=py36_0
+opencv=3.4.2=py36h6fd60c2_1
+openssl=1.1.1i=h27cfd23_0
+pcre=8.44=he6710b0_0
+pillow=8.1.0=py36he98fc37_0
+pip=20.3.3=py36h06a4308_0
+pixman=0.40.0=h7b6447c_0
+py-opencv=3.4.2=py36hb342d67_1
+pycparser=2.20=py_2
+pyparsing=2.4.7=pyhd3eb1b0_0
+pyqt=5.9.2=py36h05f1152_2
+pytables=3.4.4=py36h4f72b40_1
+python=3.6.12=hcff3b4d_2
+python-dateutil=2.8.1=py_0
+python_abi=3.6=1_cp36m
+pytorch=1.2.0=py3.6_cuda9.2.148_cudnn7.6.2_0
+pytz=2020.5=pyhd3eb1b0_0
+pywavelets=1.1.1=py36h7b6447c_2
+pyyaml=5.4.1=py36h27cfd23_1
+qt=5.9.7=h5867ecd_1
+readline=8.0=h7b6447c_0
+scikit-image=0.17.2=py36hdf5156a_0
+scikit-learn=0.23.2=py36h0573a6f_0
+scipy=1.5.2=py36h0b6359f_0
+setuptools=52.0.0=py36h06a4308_0
+sip=4.19.8=py36hf484d3e_0
+six=1.15.0=py36h06a4308_0
+sqlite=3.33.0=h62c20be_0
+threadpoolctl=2.1.0=pyh5ca1d4c_0
+tifffile=2020.10.1=py36hdd07704_2
+tk=8.6.10=hbc83047_0
+toolz=0.11.1=pyhd3eb1b0_0
+torchvision=0.4.0=py36_cu92
+tornado=6.1=py36h27cfd23_0
+tqdm=4.56.0=pyhd3eb1b0_0
+wheel=0.36.2=pyhd3eb1b0_0
+xz=5.2.5=h7b6447c_0
+yaml=0.2.5=h7b6447c_0
+zlib=1.2.11=h7b6447c_3
+zstd=1.4.5=h9ceee32_0