Self corrective Perturbations for Semantic Segmentation and Classification

by   Swami Sankaranarayanan, et al.
University of Maryland
General Electric

Convolutional Neural Networks have been a subject of great importance over the past decade and great strides have been made in their utility for producing state of the art performance in many computer vision problems. However, the behavior of deep networks is yet to be fully understood and is still an active area of research. In this work, we present an intriguing behavior: pre-trained CNNs can be made to improve their predictions by structurally perturbing the input. We observe that these perturbations - referred as Guided Perturbations - enable a trained network to improve its prediction performance without any learning or change in network weights. We perform various ablative experiments to understand how these perturbations affect the local context and feature representations. Furthermore, we demonstrate that this idea can improve performance of several existing approaches on semantic segmentation and scene labeling tasks on the PASCAL VOC dataset and supervised classification tasks on MNIST and CIFAR10 datasets.


page 3

page 4

page 6

page 11

page 12

page 13

page 14

page 15


Neither Quick Nor Proper -- Evaluation of QuickProp for Learning Deep Neural Networks

Neural networks and especially convolutional neural networks are of grea...

ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data

Scene understanding of high resolution aerial images is of great importa...

Efficient Hybrid Transformer: Learning Global-local Context for Urban Sence Segmentation

Semantic segmentation of fine-resolution urban scene images plays a vita...

Indirect Local Attacks for Context-aware Semantic Segmentation Networks

Recently, deep networks have achieved impressive semantic segmentation p...

Learning Debiased and Disentangled Representations for Semantic Segmentation

Deep neural networks are susceptible to learn biased models with entangl...

Efficient Ladder-style DenseNets for Semantic Segmentation of Large Images

Recent progress of deep image classification models has provided great p...

Please sign up or login with your details

Forgot password? Click here to reset