Efficient Transferability Assessment for Selection of Pre-trained Detectors

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Zhao, Li, Aoxue, Li, Zhenguo, Dou, Qi
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910367239634944
author Wang, Zhao
Li, Aoxue
Li, Zhenguo
Dou, Qi
author_facet Wang, Zhao
Li, Aoxue
Li, Zhenguo
Dou, Qi
contents Large-scale pre-training followed by downstream fine-tuning is an effective solution for transferring deep-learning-based models. Since finetuning all possible pre-trained models is computational costly, we aim to predict the transferability performance of these pre-trained models in a computational efficient manner. Different from previous work that seek out suitable models for downstream classification and segmentation tasks, this paper studies the efficient transferability assessment of pre-trained object detectors. To this end, we build up a detector transferability benchmark which contains a large and diverse zoo of pre-trained detectors with various architectures, source datasets and training schemes. Given this zoo, we adopt 7 target datasets from 5 diverse domains as the downstream target tasks for evaluation. Further, we propose to assess classification and regression sub-tasks simultaneously in a unified framework. Additionally, we design a complementary metric for evaluating tasks with varying objects. Experimental results demonstrate that our method outperforms other state-of-the-art approaches in assessing transferability under different target domains while efficiently reducing wall-clock time 32$\times$ and requires a mere 5.2\% memory footprint compared to brute-force fine-tuning of all pre-trained detectors.
format Preprint
id arxiv_https___arxiv_org_abs_2403_09432
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Efficient Transferability Assessment for Selection of Pre-trained Detectors
Wang, Zhao
Li, Aoxue
Li, Zhenguo
Dou, Qi
Computer Vision and Pattern Recognition
Large-scale pre-training followed by downstream fine-tuning is an effective solution for transferring deep-learning-based models. Since finetuning all possible pre-trained models is computational costly, we aim to predict the transferability performance of these pre-trained models in a computational efficient manner. Different from previous work that seek out suitable models for downstream classification and segmentation tasks, this paper studies the efficient transferability assessment of pre-trained object detectors. To this end, we build up a detector transferability benchmark which contains a large and diverse zoo of pre-trained detectors with various architectures, source datasets and training schemes. Given this zoo, we adopt 7 target datasets from 5 diverse domains as the downstream target tasks for evaluation. Further, we propose to assess classification and regression sub-tasks simultaneously in a unified framework. Additionally, we design a complementary metric for evaluating tasks with varying objects. Experimental results demonstrate that our method outperforms other state-of-the-art approaches in assessing transferability under different target domains while efficiently reducing wall-clock time 32$\times$ and requires a mere 5.2\% memory footprint compared to brute-force fine-tuning of all pre-trained detectors.
title Efficient Transferability Assessment for Selection of Pre-trained Detectors
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.09432