Feature Binding with Category-Dependant MixUp for Semantic Segmentation and Adversarial Robustness

08/13/2020
by   Md Amirul Islam, et al.
4

In this paper, we present a strategy for training convolutional neural networks to effectively resolve interference arising from competing hypotheses relating to inter-categorical information throughout the network. The premise is based on the notion of feature binding, which is defined as the process by which activation's spread across space and layers in the network are successfully integrated to arrive at a correct inference decision. In our work, this is accomplished for the task of dense image labelling by blending images based on their class labels, and then training a feature binding network, which simultaneously segments and separates the blended images. Subsequent feature denoising to suppress noisy activations reveals additional desirable properties and high degrees of successful predictions. Through this process, we reveal a general mechanism, distinct from any prior methods, for boosting the performance of the base segmentation network while simultaneously increasing robustness to adversarial attacks.

READ FULL TEXT

page 1

page 4

page 6

page 8

page 9

research
08/23/2021

SegMix: Co-occurrence Driven Mixup for Semantic Segmentation and Adversarial Robustness

In this paper, we present a strategy for training convolutional neural n...
research
04/23/2020

Improved Noise and Attack Robustness for Semantic Segmentation by Using Multi-Task Training with Self-Supervised Depth Estimation

While current approaches for neural network training often aim at improv...
research
03/24/2023

Feature Separation and Recalibration for Adversarial Robustness

Deep neural networks are susceptible to adversarial attacks due to the a...
research
05/25/2023

PEARL: Preprocessing Enhanced Adversarial Robust Learning of Image Deraining for Semantic Segmentation

In light of the significant progress made in the development and applica...
research
03/14/2022

Defending From Physically-Realizable Adversarial Attacks Through Internal Over-Activation Analysis

This work presents Z-Mask, a robust and effective strategy to improve th...
research
08/26/2022

Robust Prototypical Few-Shot Organ Segmentation with Regularized Neural-ODEs

Despite the tremendous progress made by deep learning models in image se...
research
11/20/2018

Recurrent Iterative Gating Networks for Semantic Segmentation

In this paper, we present an approach for Recurrent Iterative Gating cal...

Please sign up or login with your details

Forgot password? Click here to reset