User Constrained Thumbnail Generation Using Adaptive Convolutions
TL
TL;DR
This work proposes a deep neural framework to generate thumbnails of any size and aspect ratio, even for unseen values during training, with high accuracy and precision, over existing state-of-the-art techniques.
Abstract
Thumbnails are widely used all over the world as a preview for digital images. In this work we propose a deep neural framework to generate thumbnails of any size and aspect ratio, even for unseen values during training, with high accuracy and precision. We use Global Context Aggregation (GCA) and a modified Region Proposal Network (RPN) with adaptive convolutions to generate thumbnails in real time. GCA is used to selectively attend and aggregate the global context information from the entire image while the RPN is used to generate candidate bounding boxes for the thumbnail image. Adaptive convolution eliminates the difficulty of generating thumbnails of various aspect ratios by using filter weights dynamically generated from the aspect ratio information. The experimental results indicate the superior performance of the proposed model1 over existing state-of-the-art techniques.
References18 items
1
A Deep Network Solution for Attention and Aesthetics Aware Photo Cropping
2
A2-RL: Aesthetics Aware Reinforcement Learning for Image Cropping
3
PiCANet: Learning Pixel-Wise Contextual Attention for Saliency Detection
4
Fast-At: Fast Automatic Thumbnail Generation Using Deep Neural Networks
5