Open Set Object Recognition

Topic definitions

• open set object detection
• open world object detection
• open world tracking
• open set segmentation
• open set panoptic segmentation

List of all publications

2017

• Dropout sampling for robust object detection in open-set conditions, Dimity Miller, Lachlan Nicholson, Feras Dayoub, Niko Sunderhauf

2018

2019

2020

• The Overlooked Elephant of Object Detection: Open Set, Akshay Raj Dhamija, Manuel G ̈unther, Jonathan Ventura, and Terrance E. Boult

2021

• Exemplar-Based Open-Set Panoptic Segmentation Network, Jaedong Hwang, Seoung Wug Oh, Joon-Young Lee, Bohyung Han [CVPR 2021]
• OW-DETR: Open-world Detection Transformer Akshita Gupta, Sanath Narayan, K. J. Joseph, S. Khan, F. Khan, M. Shah
• Towards open world object detection, K J Joseph, Salman Khan, Fahad Shahbaz Khan, Vineeth N Balasubramanian
• Deep Metric Learning for Open World Semantic Segmentation, Jun Cen Peng Yun Junhao Cai Michael Yu Wang Ming Liu
• Learning open-world object proposals without learning to classify, Dahun Kim, Tsung-Yi Lin, Anelia Angelova, In So Kweon, Weicheng Kuo

2022

• Uncertainty for Identifying Open-Set Errors in Visual Object Detection Dimity Miller, N. Sunderhauf, Michael Milford, Feras Dayoub
• Learning to Detect Every Thing in an Open World, Kuniaki Saito, Ping Hu, Trevor Darrell, Kate Saenko

Breakdown by topic

Open Set Object Detection

Open World Object Detection

Open Set Segmentation

Open Set Panoptic Segmentation

Related topics

Uncertainty estimation

Benchmarks

Evaluation metrics

• Precision and Recall -
• Absolute Open Set Error - total number of observations that pass the Entropy test, fall on unknown objects (there are no overlapping groud truth objects with an IoU >= 0.5 )and do not have a winning class label of unknown. In the case all observations are known objects, open set error is 0. - definition from Miller et al

Datasets

• Scene RGB-D
• QUT Campus Dataset