Fusing Deep Convolutional Networks for Large Scale Visual Concept Classification

TL;DR

This paper investigates convolutional neural networks for large-scale visual classification, proposing efficient fusion methods that achieve state-of-the-art results with lower computational costs and minimal data augmentation.

Contribution

It introduces novel fusion mechanisms for CNNs that improve accuracy and efficiency on benchmark datasets without extensive data augmentation.

Findings

Achieved state-of-the-art results on benchmark datasets.

Proposed fusion methods reduce computational costs.

Effective performance without extensive data augmentation.

Abstract

Deep learning architectures are showing great promise in various computer vision domains including image classification, object detection, event detection and action recognition. In this study, we investigate various aspects of convolutional neural networks (CNNs) from the big data perspective. We analyze recent studies and different network architectures both in terms of running time and accuracy. We present extensive empirical information along with best practices for big data practitioners. Using these best practices we propose efficient fusion mechanisms both for single and multiple network models. We present state-of-the art results on benchmark datasets while keeping computational costs at a lower level. Another contribution of our paper is that these state-of-the-art results can be reached without using extensive data augmentation techniques.