| CARVIEW |
Select Language
HTTP/2 301
server: GitHub.com
content-type: text/html
location: https://timy90022.github.io/DropLoss/
x-github-request-id: 1D48:272D88:7C7F4E:8B7D17:69514832
accept-ranges: bytes
age: 0
date: Sun, 28 Dec 2025 15:09:38 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210020-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766934579.602978,VS0,VE201
vary: Accept-Encoding
x-fastly-request-id: 3ba3f3938a04760b2e8b68a7f2a057db6439a5e6
content-length: 162
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Wed, 14 Apr 2021 17:07:48 GMT
access-control-allow-origin: *
etag: W/"60772164-17fd"
expires: Sun, 28 Dec 2025 15:19:38 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: 68EA:1387E:7B33D0:8A33E4:69514832
accept-ranges: bytes
age: 0
date: Sun, 28 Dec 2025 15:09:39 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210020-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766934579.818140,VS0,VE223
vary: Accept-Encoding
x-fastly-request-id: 738f7fe78fbb454251b9811ab6c8e2c775c27c79
content-length: 2359
DropLoss for Long-Tail Instance Segmentation
DropLoss for Long-Tail Instance Segmentation
Ting-I Hsieh1
Esther Robb2
Hwann-Tzong Chen1,3
Jia-Bin Huang2
1National Tsing Hua University 2Virginia Tech 3Aeolus Robotics
Abstract
Long-tailed class distributions are prevalent among the practical applications of object detection and instance segmentation. Prior work in long-tail instance segmentation addresses the imbalance of losses between rare and frequent categories by reducing the penalty for a model incorrectly predicting a rare class label. We demonstrate that the rare categories are heavily suppressed by correct background predictions, which reduces the probability for all foreground categories with equal weight. Due to the relative infrequency of rare categories, this leads to an imbalance that biases towards predicting more frequent categories. Based on this insight, we develop DropLoss -- a novel adaptive loss to compensate for this imbalance without a trade-off between rare and frequent categories. With this loss, we show state-of-the-art mAP across rare, common, and frequent categories on the LVIS dataset.Citation
If you find this project useful, please cite this work:
BibTex
@inproceedings{DBLP:conf/aaai/Ting21,
author = {Ting-I Hsieh and Esther Robb and Hwann-Tzong Chen and Jia-Bin Huang},
title = {DropLoss for Long-Tail Instance Segmentation},
booktitle = {Proceedings of the Workshop on Artificial Intelligence Safety 2021
(SafeAI 2021) co-located with the Thirty-Fifth {AAAI} Conference on
Artificial Intelligence {(AAAI} 2021), Virtual, February 8, 2021},
year = {2021}
}
Copyright © Ting-I Hsieh 2020

