Toward unsupervised, multi-object discovery in large-scale image collections

This paper addresses the problem of discovering the objects present in a collection of images without any supervision. We build on the optimization approach of Vo et al. (CVPR'19) with several key novelties: (1) We propose a novel saliency-based region proposal algorithm that achieves significantly higher overlap with ground-truth objects than other competitive methods... (read more)

PDF Abstract ECCV 2020 PDF ECCV 2020 Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Multi-object discovery COCO_20k Large-scale rOSD Detection Rate 12.0 # 1
Single-object discovery COCO_20k Large-scale rOSD CorLoc 48.5 # 1
Single-object colocalization Object Discovery rOSD CorLoc 90.2 # 1
Single-object discovery Object Discovery rOSD CorLoc 89.2 # 1
Multi-object discovery VOC12 rOSD Detection Rate 40.4 # 2
Multi-object colocalization VOC12 rOSD Detection Rate 51.5 # 1
Multi-object discovery VOC12 Large-scale rOSD Detection Rate 41.2 # 1
Single-object discovery VOC12 rOSD CorLoc 51.2 # 2
Single-object colocalization VOC12 rOSD CorLoc 49.2 # 1
Single-object discovery VOC12 Large-scale rOSD CorLoc 51.9 # 1
Single-object discovery VOC_6x2 rOSD CorLoc 72.5 # 1
Single-object colocalization VOC_6x2 rOSD CorLoc 76.1 # 1
Multi-object discovery VOC_all rOSD Detection Rate 37.6 # 2
Single-object discovery VOC_all rOSD CorLoc 49.3 # 2
Single-object colocalization VOC_all rOSD CorLoc 46.7 # 1
Single-object discovery VOC_all Large-scale rOSD CorLoc 49.4 # 1
Multi-object colocalization VOC_all rOSD Detection Rate 49.4 # 1
Multi-object discovery VOC_all Large-scale rOSD Detection Rate 38.3 # 1

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet