Fix weird resized width and height if the difference of image ratio is large - #18
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Intuition
Before forward the image to model, resize the image to ensure the width and height are the multiply of stride (default: 16) is mandatory. However, there is a weird behaviour when the ratio between width and height is large.
Observation
Assume that an image in shape (40, 248, 3) is ready to be resized, the method
Predictor._resize_to_fixed_ratiois invoked and returns a resized image in shape (608, 3776), which is extremely large and improper. That may decrease the inference time without any rising performance.This behaviour is also spotted in the original code.
Solution
The solution is modify the way to compute
factor: it should bebound_dim / max(h,w).After the fix, the image is resized in shape (112, 608), which is more proper.