SOTAVerified

Long-tailed Object Detection

The training dataset obeys a long-tailed distribution, i.e., the head categories occupy most samples whereas the tail categories only own a few samples. However, the testing dataset is balanced across categories.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 25 papers

TitleStatusHype
DINO-X: A Unified Vision Model for Open-World Object Detection and UnderstandingCode5
Equalized Focal Loss for Dense Long-Tailed Object DetectionCode2
The Equalization Losses: Gradient-Driven Training for Long-tailed Object RecognitionCode2
Balanced Classification: A Unified Framework for Long-Tailed Object DetectionCode1
Long-tail Detection with Effective Class-MarginsCode1
Equalization Loss v2: A New Gradient Balance Approach for Long-tailed Object DetectionCode1
Exploring Classification Equilibrium in Long-Tailed Object DetectionCode1
MosaicOS: A Simple and Effective Use of Object-Centric Images for Long-Tailed Object DetectionCode1
Forest R-CNN: Large-Vocabulary Long-Tailed Object Detection and Instance SegmentationCode1
On Model Calibration for Long-Tailed Object Detection and Instance SegmentationCode1
Show:102550
← PrevPage 1 of 3Next →

No leaderboard results yet.