Conversation
mrn-mln
left a comment
mrn-mln
left a comment
There was a problem hiding this comment.
Great work. I reviewed the PR and left some comments.
| Name = alphapose_mobilenet_ssd | ||
| ;ImageSize should be 3 numbers seperated by commas, no spaces: 300,300,3 (for better accuracy use higher resolution when | ||
| ; using openpifpaf (openpifpaf detects both faces and pedestrians) | ||
| ImageSize = 1281,721,3 |
There was a problem hiding this comment.
The ImageSize is used for Alphapose?
I think leave a comment for ImageSize and its usage can be helpful.
| for i, box in enumerate(boxes): | ||
| inps[i], cropped_box = self._transform_single_detection(image, box) | ||
| cropped_boxes[i] = torch.FloatTensor(cropped_box) | ||
| return inps, cropped_boxes, boxes, scores, ids |
There was a problem hiding this comment.
Can you docstring the inputs and outputs shape and type.
| cropped_boxes[i] = torch.FloatTensor(cropped_box) | ||
| return inps, cropped_boxes, boxes, scores, ids | ||
|
|
||
| def _transform_single_detection(self, image, bbox): |
There was a problem hiding this comment.
Docstring of the input/output shape and type here.
|
|
||
| return img, bbox | ||
|
|
||
| def _post_process(self, hm, cropped_boxes, boxes, scores, ids): |
There was a problem hiding this comment.
Docstring of the input and output shapes and types
| @@ -0,0 +1,21 @@ | |||
| from models.fastpose import FastPose | |||
There was a problem hiding this comment.
Add the reference if you copy this module from another repo otherwise it's not neccessary.
| @@ -0,0 +1,204 @@ | |||
| import os | |||
There was a problem hiding this comment.
Is it neccessary to use this module? If not, you can remove it.
| @@ -0,0 +1,33 @@ | |||
| import numpy as np | |||
There was a problem hiding this comment.
Add reference if this module is copied from another repo.
| @@ -0,0 +1,200 @@ | |||
| import numpy as np | |||
| @@ -0,0 +1,119 @@ | |||
| import numpy as np | |||
| inference function sets input tensor to input image and gets the output. | ||
| The model provides corresponding detection output which is used for creating result | ||
| Args: | ||
| resized_rgb_image: uint8 numpy array with shape (img_height, img_width, channels) |
There was a problem hiding this comment.
change 'resized_rgb_image' to 'image' at docstring
2 participants
add AlphaPose Pose Estimator.