Saved in:
Bibliographic Details
Main Authors: Rios, Edwin Arkel, Oyerinde, Femiloye, Hu, Min-Chun, Lai, Bo-Cheng
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.11051
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929501624074240
author Rios, Edwin Arkel
Oyerinde, Femiloye
Hu, Min-Chun
Lai, Bo-Cheng
author_facet Rios, Edwin Arkel
Oyerinde, Femiloye
Hu, Min-Chun
Lai, Bo-Cheng
contents Ultra-fine-grained image recognition (UFGIR) categorizes objects with extremely small differences between classes, such as distinguishing between cultivars within the same species, as opposed to species-level classification in fine-grained image recognition (FGIR). The difficulty of this task is exacerbated due to the scarcity of samples per category. To tackle these challenges we introduce a novel approach employing down-sampling inter-layer adapters in a parameter-efficient setting, where the backbone parameters are frozen and we only fine-tune a small set of additional modules. By integrating dual-branch down-sampling, we significantly reduce the number of parameters and floating-point operations (FLOPs) required, making our method highly efficient. Comprehensive experiments on ten datasets demonstrate that our approach obtains outstanding accuracy-cost performance, highlighting its potential for practical applications in resource-constrained environments. In particular, our method increases the average accuracy by at least 6.8\% compared to other methods in the parameter-efficient setting while requiring at least 123x less trainable parameters compared to current state-of-the-art UFGIR methods and reducing the FLOPs by 30\% in average compared to other methods.
format Preprint
id arxiv_https___arxiv_org_abs_2409_11051
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Down-Sampling Inter-Layer Adapter for Parameter and Computation Efficient Ultra-Fine-Grained Image Recognition
Rios, Edwin Arkel
Oyerinde, Femiloye
Hu, Min-Chun
Lai, Bo-Cheng
Computer Vision and Pattern Recognition
I.2, I.4
Ultra-fine-grained image recognition (UFGIR) categorizes objects with extremely small differences between classes, such as distinguishing between cultivars within the same species, as opposed to species-level classification in fine-grained image recognition (FGIR). The difficulty of this task is exacerbated due to the scarcity of samples per category. To tackle these challenges we introduce a novel approach employing down-sampling inter-layer adapters in a parameter-efficient setting, where the backbone parameters are frozen and we only fine-tune a small set of additional modules. By integrating dual-branch down-sampling, we significantly reduce the number of parameters and floating-point operations (FLOPs) required, making our method highly efficient. Comprehensive experiments on ten datasets demonstrate that our approach obtains outstanding accuracy-cost performance, highlighting its potential for practical applications in resource-constrained environments. In particular, our method increases the average accuracy by at least 6.8\% compared to other methods in the parameter-efficient setting while requiring at least 123x less trainable parameters compared to current state-of-the-art UFGIR methods and reducing the FLOPs by 30\% in average compared to other methods.
title Down-Sampling Inter-Layer Adapter for Parameter and Computation Efficient Ultra-Fine-Grained Image Recognition
topic Computer Vision and Pattern Recognition
I.2, I.4
url https://arxiv.org/abs/2409.11051