# Adaptive Adjustment with Semantic Feature Space for Zero-Shot Recognition

## TL;DR

This paper introduces a novel zero-shot recognition framework that adaptively adjusts the semantic feature space to address domain shift and hubness issues, leading to improved recognition of unseen classes.

## Contribution

It is the first to propose adaptive adjustment of semantic feature space in zero-shot recognition, enhancing model robustness and training efficiency.

## Key findings

- Significant performance improvements over existing methods.
- Effective handling of domain shift and hubness problems.
- Efficient training framework for zero-shot recognition.

## Abstract

In most recent years, zero-shot recognition (ZSR) has gained increasing attention in machine learning and image processing fields. It aims at recognizing unseen class instances with knowledge transferred from seen classes. This is typically achieved by exploiting a pre-defined semantic feature space (FS), i.e., semantic attributes or word vectors, as a bridge to transfer knowledge between seen and unseen classes. However, due to the absence of unseen classes during training, the conventional ZSR easily suffers from domain shift and hubness problems. In this paper, we propose a novel ZSR learning framework that can handle these two issues well by adaptively adjusting semantic FS. To the best of our knowledge, our work is the first to consider the adaptive adjustment of semantic FS in ZSR. Moreover, our solution can be formulated to a more efficient framework that significantly boosts the training. Extensive experiments show the remarkable performance improvement of our model compared with other existing methods.