Color inference from semantic labeling for person search in videos
Published in
International Conference on Image Analysis and ...(2019)External Links:
TL
TL;DR
This work proposes an explainable model to generate semantic color labels for person search based on binary search trees and a large peer-labelled color name dataset, and uses semantic segmentation and a color labeling method to label segments of pedestrians with their associated colors.
Abstract
We propose an explainable model to generate semantic color labels for person search. In this context, persons are described from their semantic parts, such as hat, shirt, etc. Person search consists in looking for people based on these descriptions. In this work, we aim to improve the accuracy of color labels for people. Our goal is to handle the high variability of human perception. Existing solutions are based on hand-crafted features or learnt features that are not explainable. Moreover most of them only focus on a limited set of colors. We propose a method based on binary search trees and a large peer-labelled color name dataset. This allows us to synthesize the human perception of colors. Using semantic segmentation and our color labeling method, we label segments of pedestrians with their associated colors. We evaluate our solution on person search on datasets such as PCN, and show a precision as high as 80.4%.
References37 items
1
Quasi-Unsupervised Color Constancy
2
Macro-Micro Adversarial Network for Human Parsing
3
CNN Based Learning Using Reflection and Retinex Models for Intrinsic Image Decomposition
4
Improving Person Re-identification by Attribute and Identity Learning
5