Padding Creates A "blind Spot"! Examine Those Blind Spots!
3 main points
✔️ Demonstrate how padding generates CNN artifacts (blind spots)
✔️ Identify that uneven application of 0-padding is a solvable source of bias
✔️ Padding is linked to CNN's foveation behavior
Mind the Pad -- CNNs can Develop Blind Spots
written by Bilal Alsallakh, Narine Kokhlikyan, Vivek Miglani, Jun Yuan, Orion Reblitz-Richardson
(Submitted on 5 Oct 2020)
Comments: Accepted at ICLR2021
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
Paper Official Code COMM Code
Convolutional neural networks (CNNs) have become state-of-the-art feature extractors for a variety of machine learning tasks. Much work has focused on understanding the feature maps that CNN computes on input, but little attention has been paid to the spatial distribution within the feature maps. The reason for the authors' interest in this point was triggered by a mysterious case of signal detector failure. What it means is that within one frame of an on-board camera sequence, a small but visible traffic light can be detected. But in the next frame, it fails to detect the same traffic light.
It's funny. We should have been able to detect the traffic lights with high accuracy, but after only one frame, we can't. The only difference is that in the input image, the car is moving forward and so the traffic lights are shifted slightly in the vertical direction. The only difference is that in the input image, the only difference is that the signal is slightly shifted vertically because the car is moving forward. You hear a lot about CNN being invariant. Nonetheless, this research has begun because it is disconcerting that there are cases where only a slight shift makes it impossible to detect.
In conclusion, the authors attributed this to the presence of an artifact that was not a consistent feature when analyzing the spatial distribution of feature map activations (incidentally, many imaging researchers have probably seen this artifact before). In this study, we examine the causes and effects of these artifacts.
The contributions of this study are as follows
- Demonstrate how padding generates CNN artifacts
- Examine the impact of these artifacts on the task
- Identify the uneven application of 0-padding as a resolvable source of bias
- Padding is linked to CNN's foveation behavior
To read more,
Please register with AI-SCHOLAR.OR
Categories related to this article