Bidirectional Logits Tree: Pursuing Granularity Reconcilement in Fine-Grained Classification

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lu, Zhiguang, Xu, Qianqian, Bao, Shilong, Yang, Zhiyong, Huang, Qingming
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912159809667072
author Lu, Zhiguang
Xu, Qianqian
Bao, Shilong
Yang, Zhiyong
Huang, Qingming
author_facet Lu, Zhiguang
Xu, Qianqian
Bao, Shilong
Yang, Zhiyong
Huang, Qingming
contents This paper addresses the challenge of Granularity Competition in fine-grained classification tasks, which arises due to the semantic gap between multi-granularity labels. Existing approaches typically develop independent hierarchy-aware models based on shared features extracted from a common base encoder. However, because coarse-grained levels are inherently easier to learn than finer ones, the base encoder tends to prioritize coarse feature abstractions, which impedes the learning of fine-grained features. To overcome this challenge, we propose a novel framework called the Bidirectional Logits Tree (BiLT) for Granularity Reconcilement. The key idea is to develop classifiers sequentially from the finest to the coarsest granularities, rather than parallelly constructing a set of classifiers based on the same input features. In this setup, the outputs of finer-grained classifiers serve as inputs for coarser-grained ones, facilitating the flow of hierarchical semantic information across different granularities. On top of this, we further introduce an Adaptive Intra-Granularity Difference Learning (AIGDL) approach to uncover subtle semantic differences between classes within the same granularity. Extensive experiments demonstrate the effectiveness of our proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2412_12782
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Bidirectional Logits Tree: Pursuing Granularity Reconcilement in Fine-Grained Classification
Lu, Zhiguang
Xu, Qianqian
Bao, Shilong
Yang, Zhiyong
Huang, Qingming
Computer Vision and Pattern Recognition
This paper addresses the challenge of Granularity Competition in fine-grained classification tasks, which arises due to the semantic gap between multi-granularity labels. Existing approaches typically develop independent hierarchy-aware models based on shared features extracted from a common base encoder. However, because coarse-grained levels are inherently easier to learn than finer ones, the base encoder tends to prioritize coarse feature abstractions, which impedes the learning of fine-grained features. To overcome this challenge, we propose a novel framework called the Bidirectional Logits Tree (BiLT) for Granularity Reconcilement. The key idea is to develop classifiers sequentially from the finest to the coarsest granularities, rather than parallelly constructing a set of classifiers based on the same input features. In this setup, the outputs of finer-grained classifiers serve as inputs for coarser-grained ones, facilitating the flow of hierarchical semantic information across different granularities. On top of this, we further introduce an Adaptive Intra-Granularity Difference Learning (AIGDL) approach to uncover subtle semantic differences between classes within the same granularity. Extensive experiments demonstrate the effectiveness of our proposed method.
title Bidirectional Logits Tree: Pursuing Granularity Reconcilement in Fine-Grained Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.12782