Network Comparison Study of Deep Activation Feature Discriminability with Novel Objects

by   Michael Karnes, et al.

Feature extraction has always been a critical component of the computer vision field. More recently, state-of-the-art computer visions algorithms have incorporated Deep Neural Networks (DNN) in feature extracting roles, creating Deep Convolutional Activation Features (DeCAF). The transferability of DNN knowledge domains has enabled the wide use of pretrained DNN feature extraction for applications with novel object classes, especially those with limited training data. This study analyzes the general discriminability of novel object visual appearances encoded into the DeCAF space of six of the leading visual recognition DNN architectures. The results of this study characterize the Mahalanobis distances and cosine similarities between DeCAF object manifolds across two visual object tracking benchmark data sets. The backgrounds surrounding each object are also included as an object classes in the manifold analysis, providing a wider range of novel classes. This study found that different network architectures led to different network feature focuses that must to be considered in the network selection process. These results are generated from the VOT2015 and UAV123 benchmark data sets; however, the proposed methods can be applied to efficiently compare estimated network performance characteristics for any labeled visual data set.


page 3

page 4

page 5


Unsupervised Network Pretraining via Encoding Human Design

Over the years, computer vision researchers have spent an immense amount...

Learning to associate detections for real-time multiple object tracking

With the recent advances in the object detection research field, trackin...

DeepRare: Generic Unsupervised Visual Attention Models

Human visual system is modeled in engineering field providing feature-en...

DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition

We evaluate whether features extracted from the activation of a deep con...

Efficient Scale Estimation Methods using Lightweight Deep Convolutional Neural Networks for Visual Tracking

In recent years, visual tracking methods that are based on discriminativ...

Deep Active Audio Feature Learning in Resource-Constrained Environments

The scarcity of labelled data makes training Deep Neural Network (DNN) m...

Visual link retrieval and knowledge discovery in painting datasets

Visual arts have invaluable importance for the cultural, historic and ec...

Please sign up or login with your details

Forgot password? Click here to reset