AI Chat AI Image Generator AI Video Text to Speech

What Do Deep CNNs Learn About Objects?

04/09/2015

∙

by Xingchao Peng, et al.

∙

∙

Deep convolutional neural networks learn extremely powerful image representations, yet most of that power is hidden in the millions of deep-layer parameters. What exactly do these parameters represent? Recent work has started to analyse CNN representations, finding that, e.g., they are invariant to some 2D transformations Fischer et al. (2014), but are confused by particular types of image noise Nguyen et al. (2014). In this work, we delve deeper and ask: how invariant are CNNs to object-class variations caused by 3D shape, pose, and photorealism?

Xingchao Peng
14 publications
Baochen Sun
8 publications
Karim Ali
6 publications
Kate Saenko
150 publications

research

∙ 03/21/2015

Boosting Convolutional Features for Robust Object Proposals

Deep Convolutional Neural Networks (CNNs) have demonstrated excellent pe...

0 Nikolaos Karianakis, et al. ∙

research

∙ 03/15/2018

Studying Invariances of Trained Convolutional Neural Networks

Convolutional Neural Networks (CNNs) define an exceptionally powerful cl...

0 Charlotte Bunne, et al. ∙

research

∙ 08/31/2016

Convolutional Neural Networks for Text Categorization: Shallow Word-level vs. Deep Character-level

This paper reports the performances of shallow word-level convolutional ...

0 Rie Johnson, et al. ∙

research

∙ 05/30/2018

Why do deep convolutional networks generalize so poorly to small image transformations?

Deep convolutional network architectures are often assumed to guarantee ...

4 Aharon Azulay, et al. ∙

research

∙ 11/19/2018

Deeper Interpretability of Deep Networks

Deep Convolutional Neural Networks (CNNs) have been one of the most infl...

14 Tian Xu, et al. ∙

research

∙ 06/08/2022

Neural Collapse: A Review on Modelling Principles and Generalization

With a recent observation of the "Neural Collapse (NC)" phenomena by Pap...

1 Vignesh Kothapalli, et al. ∙

research

∙ 02/06/2023

Spatial Functa: Scaling Functa to ImageNet Classification and Generation

Neural fields, also known as implicit neural representations, have emerg...

0 Matthias Bauer, et al. ∙