Diving Deep into Sentiment: Understanding Fine-tuned CNNs for Visual Sentiment Prediction (2015-08-20T00:00:00.000000Z)

TL;DR

This work study the suitability of fine-tuning a CNN for visual sentiment prediction as well as explore performance boosting techniques within this deep learning setting and provides a deep-dive analysis into a benchmark, state-of-the-art network architecture to gain insight about how to design patterns for CNNs on the task ofVisual sentiment prediction.

Abstract

Visual media are powerful means of expressing emotions and sentiments. The constant generation of new content in social networks highlights the need of automated visual sentiment analysis tools. While Convolutional Neural Networks (CNNs) have established a new state-of-the-art in several vision problems, their application to the task of sentiment analysis is mostly unexplored and there are few studies regarding how to design CNNs for this purpose. In this work, we study the suitability of fine-tuning a CNN for visual sentiment prediction as well as explore performance boosting techniques within this deep learning setting. Finally, we provide a deep-dive analysis into a benchmark, state-of-the-art network architecture to gain insight about how to design patterns for CNNs on the task of visual sentiment prediction.

Authors

Xavier Giró-i-Nieto

12 papers

Amaia Salvador

6 papers

Brendan Jou

3 papers

Diving Deep into Sentiment: Understanding Fine-tuned CNNs for Visual Sentiment Prediction

TL;DR

Abstract

Authors

References32 items

A mixed bag of emotions: Model, predict, and transfer emotion distributions

Cultural Event recognition with visual ConvNets and temporal models

Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

Robust Image Sentiment Analysis Using Progressively Trained and Domain Transferred Deep Networks

Object Detectors Emerge in Deep Scene CNNs

Predicting Ad Liking and Purchase Intent: Large-Scale Analysis of Facial Responses to Ads

Visual Sentiment Prediction with Deep Convolutional Neural Networks

Predicting Viewer Perceived Emotions in Animated GIFs

A framework of changing image emotion using emotion prediction

Going deeper with convolutions

Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks

Caffe: Convolutional Architecture for Fast Feature Embedding

Predicting Emotions in User-Generated Videos

Return of the Devil in the Details: Delving Deep into Convolutional Nets

CNN Features Off-the-Shelf: An Astounding Baseline for Recognition

Intriguing properties of neural networks

Visualizing and Understanding Convolutional Networks

Affective image adjustment with a single word

Large-scale visual sentiment ontology and detectors using adjective noun pairs

Deep Learning using Linear Support Vector Machines

Deep learning for robust feature generation in audiovisual emotion recognition

ImageNet classification with deep convolutional neural networks

Can we understand van gogh's mood?: learning to infer affects from images in social networks

Common Sense Reasoning for Detection, Prevention, and Mitigation of Cyberbullying

Analyzing and predicting sentiment of images on the social web

The wisdom of social multimedia: using flickr for prediction and forecast

Affective image classification using features inspired by psychology and art theory

ImageNet: A large-scale hierarchical image database

Emotional valence categorization using holistic image features

Gradient-based learning applied to document recognition

International Affective Picture System (IAPS) : Technical Manual and Affective Ratings

Emotion, a psychoevolutionary synthesis

Field of Study

Journal Information

Name