From 04c86c8466470406d0b0f716fb4a368b832a499d Mon Sep 17 00:00:00 2001
From: ageorge <anjith.george@idiap.ch>
Date: Fri, 3 May 2019 15:01:18 +0200
Subject: [PATCH] Paths to models
---
.../mccnn/tifs2018/config/FASNet_config.py | 98 +++++++++----------
.../mccnn/tifs2018/config/MCCNN_config.py | 93 +++++++++---------
2 files changed, 91 insertions(+), 100 deletions(-)
diff --git a/bob/paper/mccnn/tifs2018/config/FASNet_config.py b/bob/paper/mccnn/tifs2018/config/FASNet_config.py
index 2acbff1..861bbd3 100644
--- a/bob/paper/mccnn/tifs2018/config/FASNet_config.py
+++ b/bob/paper/mccnn/tifs2018/config/FASNet_config.py
@@ -18,7 +18,7 @@ from torchvision import transforms
from bob.learn.pytorch.datasets import ChannelSelect
# names of the channels to process:
-_channel_names = ['color','depth','infrared','thermal']
+_channel_names = ['color', 'depth', 'infrared', 'thermal']
# dictionary containing preprocessors for all channels:
_preprocessors = {}
@@ -34,18 +34,18 @@ FACE_DETECTION_METHOD = 'mtcnn' # use ANNOTATIONS
MIN_FACE_SIZE = 50 # skip small faces
ALIGNMENT_TYPE = 'lightcnn'
-_image_preprocessor = FaceCropAlign(face_size = FACE_SIZE,
- rgb_output_flag = RGB_OUTPUT_FLAG,
- use_face_alignment = USE_FACE_ALIGNMENT,
- alignment_type =ALIGNMENT_TYPE,
- max_image_size = MAX_IMAGE_SIZE,
- face_detection_method = FACE_DETECTION_METHOD,
- min_face_size = MIN_FACE_SIZE)
+_image_preprocessor = FaceCropAlign(face_size=FACE_SIZE,
+ rgb_output_flag=RGB_OUTPUT_FLAG,
+ use_face_alignment=USE_FACE_ALIGNMENT,
+ alignment_type=ALIGNMENT_TYPE,
+ max_image_size=MAX_IMAGE_SIZE,
+ face_detection_method=FACE_DETECTION_METHOD,
+ min_face_size=MIN_FACE_SIZE)
-_frame_selector = FrameSelector(selection_style = "all")
+_frame_selector = FrameSelector(selection_style="all")
-_preprocessor_rgb = Wrapper(preprocessor = _image_preprocessor,
- frame_selector = _frame_selector)
+_preprocessor_rgb = Wrapper(preprocessor=_image_preprocessor,
+ frame_selector=_frame_selector)
_preprocessors[_channel_names[0]] = _preprocessor_rgb
@@ -60,20 +60,20 @@ FACE_DETECTION_METHOD = None # use annotations
MIN_FACE_SIZE = 50 # skip small faces
NORMALIZATION_FUNCTION = _norm_func
NORMALIZATION_FUNCTION_KWARGS = {}
-NORMALIZATION_FUNCTION_KWARGS = {'n_sigma':3.0, 'norm_method':'MAD'}
+NORMALIZATION_FUNCTION_KWARGS = {'n_sigma': 3.0, 'norm_method': 'MAD'}
-_image_preprocessor_ir = FaceCropAlign(face_size = FACE_SIZE,
- rgb_output_flag = RGB_OUTPUT_FLAG,
- use_face_alignment = USE_FACE_ALIGNMENT,
- alignment_type =ALIGNMENT_TYPE,
- max_image_size = MAX_IMAGE_SIZE,
- face_detection_method = FACE_DETECTION_METHOD,
- min_face_size = MIN_FACE_SIZE,
- normalization_function = NORMALIZATION_FUNCTION,
- normalization_function_kwargs = NORMALIZATION_FUNCTION_KWARGS)
+_image_preprocessor_ir = FaceCropAlign(face_size=FACE_SIZE,
+ rgb_output_flag=RGB_OUTPUT_FLAG,
+ use_face_alignment=USE_FACE_ALIGNMENT,
+ alignment_type=ALIGNMENT_TYPE,
+ max_image_size=MAX_IMAGE_SIZE,
+ face_detection_method=FACE_DETECTION_METHOD,
+ min_face_size=MIN_FACE_SIZE,
+ normalization_function=NORMALIZATION_FUNCTION,
+ normalization_function_kwargs=NORMALIZATION_FUNCTION_KWARGS)
-_preprocessor_ir = Wrapper(preprocessor = _image_preprocessor_ir,
- frame_selector = _frame_selector)
+_preprocessor_ir = Wrapper(preprocessor=_image_preprocessor_ir,
+ frame_selector=_frame_selector)
_preprocessors[_channel_names[1]] = _preprocessor_ir
@@ -83,9 +83,9 @@ _preprocessors[_channel_names[2]] = _preprocessor_ir
_preprocessors[_channel_names[3]] = _preprocessor_ir
-preprocessor = VideoFaceCropAlignBlockPatch(preprocessors = _preprocessors,
- channel_names = _channel_names,
- return_multi_channel_flag = True)
+preprocessor = VideoFaceCropAlignBlockPatch(preprocessors=_preprocessors,
+ channel_names=_channel_names,
+ return_multi_channel_flag=True)
#====================================================================================
@@ -96,42 +96,31 @@ from bob.learn.pytorch.extractor.image import FASNetExtractor
from bob.bio.video.extractor import Wrapper
-# MODEL_FILE= None # Replace with '<PATH_TO_MODEL>'
+# MODEL_FILE= None # Replace with '<PATH_TO_MODEL>'
# ####################################################################
# If you want to use the pretrained model
import pkg_resources
-MODEL_FILE = pkg_resources.resource_filename( 'bob.paper.mccnn.tifs2018', 'models/fasnet.pth')
+MODEL_FILE = pkg_resources.resource_filename('bob.paper.mccnn.tifs2018', 'models/fasnet.pth')
-URL='http://www.idiap.ch/~ageorge/model_100_0.pth'
+URL = 'https://www.idiap.ch/software/bob/data/bob/bob.paper.mccnn.tifs2018/master/fasnet.pth'
if not os.path.exists(MODEL_FILE):
- logger.info('Downloading the FASNet model')
+ logger.info('Downloading the FASNet model')
- bob.extension.download.download_file(URL,MODEL_FILE)
+ bob.extension.download.download_file(URL, MODEL_FILE)
- logger.info('Downloaded FASNet model to location: {}'.format(MODEL_FILE))
+ logger.info('Downloaded FASNet model to location: {}'.format(MODEL_FILE))
+_img_transform = transforms.Compose([transforms.ToPILImage(), transforms.Resize(224, interpolation=2), transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406],
+ std=[0.229, 0.224, 0.225])])
-
-MODEL_FILE=
-
-/idiap/temp/ageorge/MCCNN_paperpackage/bob.paper.mccnn.tifs2018/src/bob.paper.mccnn.tifs2018/bob/paper/mccnn/tifs2018/models/mccnn_best_C1-B1-FFC.pth
-/idiap/temp/ageorge/MCCNN_paperpackage/bob.paper.mccnn.tifs2018/src/bob.paper.mccnn.tifs2018/bob/paper/mccnn/tifs2018/models/fasnet.pth
-
-
-
-
-_img_transform = transforms.Compose([transforms.ToPILImage(),transforms.Resize(224, interpolation=2),transforms.ToTensor(),transforms.Normalize(mean=[0.485, 0.456, 0.406],
- std=[0.229, 0.224, 0.225])])
-
-
-_image_extracor=FASNetExtractor(transforms=_img_transform, model_file=MODEL_FILE)
+_image_extracor = FASNetExtractor(transforms=_img_transform, model_file=MODEL_FILE)
extractor = Wrapper(_image_extracor)
@@ -141,26 +130,27 @@ extractor = Wrapper(_image_extracor)
from bob.pad.base.algorithm import Algorithm
+
class DummyAlgorithm(Algorithm):
"""An algorithm that takes the precomputed predictions and uses them for
scoring."""
def __init__(self, **kwargs):
- super(DummyAlgorithm, self).__init__(
- **kwargs)
+ super(DummyAlgorithm, self).__init__(
+ **kwargs)
def project(self, feature):
- # print("feature",feature.as_array())
- return feature.as_array().reshape(-1,1)
-
+ # print("feature",feature.as_array())
+ return feature.as_array().reshape(-1, 1)
def score_for_multiple_projections(self, predictions):
- # one node at the output
+ # one node at the output
- return list(predictions)
+ return list(predictions)
def score(self, predictions):
- return list(predictions)
+ return list(predictions)
+
algorithm = DummyAlgorithm(performs_projection=True, requires_projector_training=False)
diff --git a/bob/paper/mccnn/tifs2018/config/MCCNN_config.py b/bob/paper/mccnn/tifs2018/config/MCCNN_config.py
index d698bcc..b36b5b7 100644
--- a/bob/paper/mccnn/tifs2018/config/MCCNN_config.py
+++ b/bob/paper/mccnn/tifs2018/config/MCCNN_config.py
@@ -16,7 +16,7 @@ from torchvision import transforms
from bob.learn.pytorch.datasets import ChannelSelect
# names of the channels to process:
-_channel_names = ['color','depth','infrared','thermal']
+_channel_names = ['color', 'depth', 'infrared', 'thermal']
# dictionary containing preprocessors for all channels:
_preprocessors = {}
@@ -32,18 +32,18 @@ FACE_DETECTION_METHOD = 'mtcnn' # use ANNOTATIONS
MIN_FACE_SIZE = 50 # skip small faces
ALIGNMENT_TYPE = 'lightcnn'
-_image_preprocessor = FaceCropAlign(face_size = FACE_SIZE,
- rgb_output_flag = RGB_OUTPUT_FLAG,
- use_face_alignment = USE_FACE_ALIGNMENT,
- alignment_type =ALIGNMENT_TYPE,
- max_image_size = MAX_IMAGE_SIZE,
- face_detection_method = FACE_DETECTION_METHOD,
- min_face_size = MIN_FACE_SIZE)
+_image_preprocessor = FaceCropAlign(face_size=FACE_SIZE,
+ rgb_output_flag=RGB_OUTPUT_FLAG,
+ use_face_alignment=USE_FACE_ALIGNMENT,
+ alignment_type=ALIGNMENT_TYPE,
+ max_image_size=MAX_IMAGE_SIZE,
+ face_detection_method=FACE_DETECTION_METHOD,
+ min_face_size=MIN_FACE_SIZE)
-_frame_selector = FrameSelector(selection_style = "all")
+_frame_selector = FrameSelector(selection_style="all")
-_preprocessor_rgb = Wrapper(preprocessor = _image_preprocessor,
- frame_selector = _frame_selector)
+_preprocessor_rgb = Wrapper(preprocessor=_image_preprocessor,
+ frame_selector=_frame_selector)
_preprocessors[_channel_names[0]] = _preprocessor_rgb
@@ -58,20 +58,20 @@ FACE_DETECTION_METHOD = None # use annotations
MIN_FACE_SIZE = 50 # skip small faces
NORMALIZATION_FUNCTION = _norm_func
NORMALIZATION_FUNCTION_KWARGS = {}
-NORMALIZATION_FUNCTION_KWARGS = {'n_sigma':3.0, 'norm_method':'MAD'}
+NORMALIZATION_FUNCTION_KWARGS = {'n_sigma': 3.0, 'norm_method': 'MAD'}
-_image_preprocessor_ir = FaceCropAlign(face_size = FACE_SIZE,
- rgb_output_flag = RGB_OUTPUT_FLAG,
- use_face_alignment = USE_FACE_ALIGNMENT,
- alignment_type =ALIGNMENT_TYPE,
- max_image_size = MAX_IMAGE_SIZE,
- face_detection_method = FACE_DETECTION_METHOD,
- min_face_size = MIN_FACE_SIZE,
- normalization_function = NORMALIZATION_FUNCTION,
- normalization_function_kwargs = NORMALIZATION_FUNCTION_KWARGS)
+_image_preprocessor_ir = FaceCropAlign(face_size=FACE_SIZE,
+ rgb_output_flag=RGB_OUTPUT_FLAG,
+ use_face_alignment=USE_FACE_ALIGNMENT,
+ alignment_type=ALIGNMENT_TYPE,
+ max_image_size=MAX_IMAGE_SIZE,
+ face_detection_method=FACE_DETECTION_METHOD,
+ min_face_size=MIN_FACE_SIZE,
+ normalization_function=NORMALIZATION_FUNCTION,
+ normalization_function_kwargs=NORMALIZATION_FUNCTION_KWARGS)
-_preprocessor_ir = Wrapper(preprocessor = _image_preprocessor_ir,
- frame_selector = _frame_selector)
+_preprocessor_ir = Wrapper(preprocessor=_image_preprocessor_ir,
+ frame_selector=_frame_selector)
_preprocessors[_channel_names[1]] = _preprocessor_ir
@@ -80,9 +80,9 @@ _preprocessors[_channel_names[2]] = _preprocessor_ir
_preprocessors[_channel_names[3]] = _preprocessor_ir
-preprocessor = VideoFaceCropAlignBlockPatch(preprocessors = _preprocessors,
- channel_names = _channel_names,
- return_multi_channel_flag = True)
+preprocessor = VideoFaceCropAlignBlockPatch(preprocessors=_preprocessors,
+ channel_names=_channel_names,
+ return_multi_channel_flag=True)
#====================================================================================
@@ -92,38 +92,38 @@ from bob.learn.pytorch.extractor.image import MCCNNExtractor
from bob.bio.video.extractor import Wrapper
-# MODEL_FILE= None # Replace with '<PATH_TO_MODEL>'
+# MODEL_FILE= None # Replace with '<PATH_TO_MODEL>'
# ####################################################################
# If you want to use the pretrained model
import pkg_resources
-MODEL_FILE = pkg_resources.resource_filename( 'bob.paper.mccnn.tifs2018', 'models/mccnn_best_C1-B1-FFC.pth')
+MODEL_FILE = pkg_resources.resource_filename('bob.paper.mccnn.tifs2018', 'models/mccnn_best_C1-B1-FFC.pth')
-URL='http://www.idiap.ch/~ageorge/model_100_0.pth'
+URL = 'http://www.idiap.ch/software/bob/data/bob/bob.paper.mccnn.tifs2018/master/mccnn_best_C1-B1-FFC.pth'
if not os.path.exists(MODEL_FILE):
- logger.info('Downloading the MCCNN model')
+ logger.info('Downloading the MCCNN model')
- bob.extension.download.download_file(URL,MODEL_FILE)
+ bob.extension.download.download_file(URL, MODEL_FILE)
- logger.info('Downloaded MCCNN model to location: {}'.format(MODEL_FILE))
+ logger.info('Downloaded MCCNN model to location: {}'.format(MODEL_FILE))
-ADAPTED_LAYERS= 'conv1-group1-ffc'
+ADAPTED_LAYERS = 'conv1-group1-ffc'
####################################################################
-SELECTED_CHANNELS= [0,1,2,3]
+SELECTED_CHANNELS = [0, 1, 2, 3]
####################################################################
-NUM_CHANNELS_USED=len(SELECTED_CHANNELS)
+NUM_CHANNELS_USED = len(SELECTED_CHANNELS)
-_img_transform = transforms.Compose([ChannelSelect(selected_channels = SELECTED_CHANNELS),transforms.ToTensor()])
+_img_transform = transforms.Compose([ChannelSelect(selected_channels=SELECTED_CHANNELS), transforms.ToTensor()])
-_image_extracor=MCCNNExtractor(num_channels_used=NUM_CHANNELS_USED, transforms=_img_transform, model_file=MODEL_FILE)
+_image_extracor = MCCNNExtractor(num_channels_used=NUM_CHANNELS_USED, transforms=_img_transform, model_file=MODEL_FILE)
extractor = Wrapper(_image_extracor)
@@ -133,26 +133,27 @@ extractor = Wrapper(_image_extracor)
from bob.pad.base.algorithm import Algorithm
+
class DummyAlgorithm(Algorithm):
"""An algorithm that takes the precomputed predictions and uses them for
scoring."""
def __init__(self, **kwargs):
- super(DummyAlgorithm, self).__init__(
- **kwargs)
+ super(DummyAlgorithm, self).__init__(
+ **kwargs)
def project(self, feature):
- # print("feature",feature.as_array())
- return feature.as_array().reshape(-1,1)
-
+ # print("feature",feature.as_array())
+ return feature.as_array().reshape(-1, 1)
def score_for_multiple_projections(self, predictions):
- # one node at the output
+ # one node at the output
- return list(predictions)
+ return list(predictions)
def score(self, predictions):
- return list(predictions)
+ return list(predictions)
+
-algorithm = DummyAlgorithm(performs_projection=True, requires_projector_training=False)
\ No newline at end of file
+algorithm = DummyAlgorithm(performs_projection=True, requires_projector_training=False)
--
GitLab