DeepAI AI Chat
Log In Sign Up

Deep Visual City Recognition Visualization

by   Xiangwei Shi, et al.

Understanding how cities visually differ from each others is interesting for planners, residents, and historians. We investigate the interpretation of deep features learned by convolutional neural networks (CNNs) for city recognition. Given a trained city recognition network, we first generate weighted masks using the known Grad-CAM technique and to select the most discriminate regions in the image. Since the image classification label is the city name, it contains no information of objects that are class-discriminate, we investigate the interpretability of deep representations with two methods. (i) Unsupervised method is used to cluster the objects appearing in the visual explanations. (ii) A pretrained semantic segmentation model is used to label objects in pixel level, and then we introduce statistical measures to quantitatively evaluate the interpretability of discriminate objects. The influence of network architectures and random initializations in training, is studied on the interpretability of CNN features for city recognition. The results suggest that network architectures would affect the interpretability of learned visual representations greater than different initializations.


page 1

page 2

page 3

page 4

page 5


Interpreting Deep Visual Representations via Network Dissection

The success of recent deep convolutional neural networks (CNNs) depends ...

Understanding Deep Architectures by Interpretable Visual Summaries

A consistent body of research investigates the recurrent visual patterns...

Visual Interpretability for Deep Learning: a Survey

This paper reviews recent studies in emerging directions of understandin...

PCACE: A Statistical Approach to Ranking Neurons for CNN Interpretability

In this paper we introduce a new problem within the growing literature o...

Interpreting Adversarial Examples by Activation Promotion and Suppression

It is widely known that convolutional neural networks (CNNs) are vulnera...

TAX: Tendency-and-Assignment Explainer for Semantic Segmentation with Multi-Annotators

To understand how deep neural networks perform classification prediction...

An Analysis of Human-centered Geolocation

Online social networks contain a constantly increasing amount of images ...

Code Repositories


Visual City Recognition

view repo