Search and Detect: Training-Free Long Tail Object Detection via Web-Image Retrieval

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sidhu, Mankeerat, Chopra, Hetarth, Blume, Ansel, Kim, Jeonghwan, Reddy, Revanth Gangi, Ji, Heng
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929518201012224
author Sidhu, Mankeerat
Chopra, Hetarth
Blume, Ansel
Kim, Jeonghwan
Reddy, Revanth Gangi
Ji, Heng
author_facet Sidhu, Mankeerat
Chopra, Hetarth
Blume, Ansel
Kim, Jeonghwan
Reddy, Revanth Gangi
Ji, Heng
contents In this paper, we introduce SearchDet, a training-free long-tail object detection framework that significantly enhances open-vocabulary object detection performance. SearchDet retrieves a set of positive and negative images of an object to ground, embeds these images, and computes an input image-weighted query which is used to detect the desired concept in the image. Our proposed method is simple and training-free, yet achieves over 48.7% mAP improvement on ODinW and 59.1% mAP improvement on LVIS compared to state-of-the-art models such as GroundingDINO. We further show that our approach of basing object detection on a set of Web-retrieved exemplars is stable with respect to variations in the exemplars, suggesting a path towards eliminating costly data annotation and training procedures.
format Preprint
id arxiv_https___arxiv_org_abs_2409_18733
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Search and Detect: Training-Free Long Tail Object Detection via Web-Image Retrieval
Sidhu, Mankeerat
Chopra, Hetarth
Blume, Ansel
Kim, Jeonghwan
Reddy, Revanth Gangi
Ji, Heng
Computer Vision and Pattern Recognition
In this paper, we introduce SearchDet, a training-free long-tail object detection framework that significantly enhances open-vocabulary object detection performance. SearchDet retrieves a set of positive and negative images of an object to ground, embeds these images, and computes an input image-weighted query which is used to detect the desired concept in the image. Our proposed method is simple and training-free, yet achieves over 48.7% mAP improvement on ODinW and 59.1% mAP improvement on LVIS compared to state-of-the-art models such as GroundingDINO. We further show that our approach of basing object detection on a set of Web-retrieved exemplars is stable with respect to variations in the exemplars, suggesting a path towards eliminating costly data annotation and training procedures.
title Search and Detect: Training-Free Long Tail Object Detection via Web-Image Retrieval
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.18733