SOTAVerified

Novel Object Detection

Novel Object Detection is a challenging task introduced by Fomenko et.al. in their paper "Learning to Discover and Detect Objects". The goal in this task is to measure mAP performance on known as well as novel classes, where the known classes correspond to the 80 COCO classes, and the novel classes are the remaining 1123 classes from LVIS dataset. Thus, during training the model can only be trained with annotations from COCO dataset, but during evaluation/inference it is expected to BOTH classify and detect objects belonging to ALL the classes in the LVIS dataset.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 31–40 of 53 papers

TitleStatusHype
PV-RCNN++: Semantical Point-Voxel Feature Interaction for 3D Object Detection—0
Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection—0
Small Instance Detection by Integer Programming on Object Density Maps—0
Unsupervised Discovery of the Long-Tail in Instance Segmentation Using Hierarchical Self-Supervision—0
Visual Understanding of Complex Table Structures from Document Images—0
Accurate Object Detection with Joint Classification-Regression Random Forests—0
LiDAR Cluster First and Camera Inference Later: A New Perspective Towards Autonomous Driving—0
MambaNeXt-YOLO: A Hybrid State Space Model for Real-time Object Detection—0
MASF-YOLO: An Improved YOLOv11 Network for Small Object Detection on Drone View—0
Meta-tuning Loss Functions and Data Augmentation for Few-shot Object Detection—0
Show:102550
← PrevPage 4 of 6Next →

No leaderboard results yet.