-
Pascal VOC-OS
Open-set object detection datasets -
COCO panoptic validation set
Panoptic segmentation aims to unify instance and semantic segmentation in the same framework. Existing works propose to merge instance and semantic segmentation using... -
COCO panoptic segmentation
Panoptic segmentation aims to unify instance and semantic segmentation in the same framework. Existing works propose to merge instance and semantic segmentation using... -
Distilling Object Detectors with Feature Richness
The proposed Feature-Richness Score (FRS) method to choose important features that are beneficial to distillation. -
CamVid Dataset
CamVid dataset is a benchmark dataset for semantic segmentation. It consists of 700 images with 11 object classes. -
Stanford Cars dataset
The Stanford Cars dataset is a dataset of images of cars, with 196 categories and approximately 16,000 images. The authors created a synthetic dataset by adding occlusions of... -
GPS-Assisted Automatic Object Annotation in Videos
TagMe is a system that provides automatic object annotation in videos using GPS data. -
Inter-Instance Similarity Modeling for Contrastive Learning
The existing contrastive learning methods widely adopt one-hot instance discrimination as pretext task for self-supervised learning, which inevitably neglects rich... -
Deformable DETR dataset
The Deformable DETR dataset -
DAVIS-2017
The DAVIS-2017 dataset is a benchmark for video object segmentation -
Light-Weight RetinaNet for Object Detection
Object detection has gained great progress driven by the development of deep learning. Compared with a widely studied task – classification, generally speaking, object detection... -
Google Street View Road Damage Dataset
Road damage data collected from Google Street View for training deep learning models. -
Crowdsensing-based Road Damage Detection Challenge (CRDDC2022)
Roadway damage data collection using Google Street View and YOLOv7 for automatic road damage detection and classification. -
KITTI Object Detection Benchmark
The KITTI Object Detection Benchmark consists of 7,481 training images and 7,518 testing images, with 3D LiDAR point clouds and camera images. -
CenterPoseTrack
CenterPoseTrack is a joint detection and tracking method for category-level 6-DoF pose estimation of previously unseen object instances.
