A benchmark of CNN backbones on DINO-DETR performance in object detection

A benchmark of CNN backbones on DINO-DETR performance in object detection

Recent developments in DETR-based models have made significant improvements in training convergence but not small object detection. This paper combines the ConvNeXt and FocalNet backbones with DINO-DETR using timm and detrex, and presents a benchmark and analysis of the resulting model performances...

Cijeli opis

Bibliografski detalji
Glavni autor: Liew, Zon Hur Zhen
Daljnji autori: Lu Shijian
Format: Final Year Project (FYP)
Jezik:English
Izdano: Nanyang Technological University 2023
Teme:
Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Online pristup:https://hdl.handle.net/10356/172020

Internet

https://hdl.handle.net/10356/172020

Slični predmeti