DyRA：面向现有检测器的可移植动态分辨率调整网络

This paper has been withdrawn by Daeun Seo

No PDF available, click to view other formats

Abstract:Achieving constant accuracy in object detection is challenging due to the inherent variability of object sizes. One effective approach to this problem involves optimizing input resolution, referred to as a multi-resolution strategy. Previous approaches to resolution optimization have often been based on pre-defined resolutions with manual selection. However, there is a lack of study on run-time resolution optimization for existing architectures. This paper introduces DyRA, a dynamic resolution adjustment network providing an image-specific scale factor for existing detectors. This network is co-trained with detectors utilizing specially designed loss functions, namely ParetoScaleLoss and BalanceLoss. ParetoScaleLoss determines an adaptive scale factor for robustness, while BalanceLoss optimizes overall scale factors according to the localization performance of the detector. The loss function is devised to minimize the accuracy drop across contrasting objectives of different-sized objects for scaling. Our proposed network can improve accuracy across various models, including RetinaNet, Faster-RCNN, FCOS, DINO, and H-Deformable-DETR. The code is available at this https URL.

Comments:	The formulation for scaling contains unnecessary parts or marginal effects on accuracy. See our recent paper, which reformulates loss functions and corresponding explanations
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2311.17098 [cs.CV]
	(or arXiv:2311.17098v4 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2311.17098 arXiv-issued DOI via DataCite

Submission history

From: Daeun Seo [view email]
[v1] Tue, 28 Nov 2023 07:52:41 UTC (2,124 KB)
[v2] Thu, 7 Dec 2023 12:33:17 UTC (2,133 KB)
[v3] Thu, 14 Mar 2024 13:22:51 UTC (6,483 KB)
[v4] Tue, 27 Oct 2026 07:00:26 UTC (1 KB) (withdrawn)