Semantic Diversity Learning for Zero-Shot Multi-label Classification

05/12/2021
by   Avi Ben-Cohen, et al.
0

Training a neural network model for recognizing multiple labels associated with an image, including identifying unseen labels, is challenging, especially for images that portray numerous semantically diverse labels. As challenging as this task is, it is an essential task to tackle since it represents many real-world cases, such as image retrieval of natural images. We argue that using a single embedding vector to represent an image, as commonly practiced, is not sufficient to rank both relevant seen and unseen labels accurately. This study introduces an end-to-end model training for multi-label zero-shot learning that supports semantic diversity of the images and labels. We propose to use an embedding matrix having principal embedding vectors trained using a tailored loss function. In addition, during training, we suggest up-weighting in the loss function image samples presenting higher semantic diversity to encourage the diversity of the embedding matrix. Extensive experiments show that our proposed method improves the zero-shot model's quality in tag-based image retrieval achieving SoTA results on several common datasets (NUS-Wide, COCO, Open Images).

READ FULL TEXT

page 7

page 12

research
12/22/2015

Multi-Instance Visual-Semantic Embedding

Visual-semantic embedding models have been recently proposed and shown t...
research
11/17/2019

Transductive Zero-Shot Hashing for Multi-Label Image Retrieval

Hash coding has been widely used in approximate nearest neighbor search ...
research
04/12/2022

Three-Stream Joint Network for Zero-Shot Sketch-Based Image Retrieval

The Zero-Shot Sketch-based Image Retrieval (ZS-SBIR) is a challenging ta...
research
03/16/2018

Deep Multiple Instance Learning for Zero-shot Image Tagging

In-line with the success of deep learning on traditional recognition pro...
research
06/12/2023

Augmenting Zero-Shot Detection Training with Image Labels

Zero-shot detection (ZSD), i.e., detection on classes not seen during tr...
research
05/12/2022

Open Vocabulary Extreme Classification Using Generative Models

The extreme multi-label classification (XMC) task aims at tagging conten...
research
03/08/2020

Unifying Specialist Image Embedding into Universal Image Embedding

Deep image embedding provides a way to measure the semantic similarity o...

Please sign up or login with your details

Forgot password? Click here to reset