Lune

ACM MM2021Top-tier venue

Towards Accurate Localization by Instance Search

Yi-Geng Hong, Hui-Chu Xiao, Wan-Lei Zhao

Abstract

Visual object localization is the key step in a series of object detection tasks. In the literature, high localization accuracy is achieved with the mainstream strongly supervised frameworks. However, such methods require object-level annotations and are unable to detect objects of unknown categories. Weakly supervised methods face similar difficulties. In this paper, a self-paced learning framework is proposed to achieve accurate object localization on the rank list returned by instance search. The proposed framework mines the target instance gradually from the queries and their corresponding top-ranked search results. Since a common instance is shared between the query and the images in the rank list, the target visual instance can be accurately localized even without knowing what the object category is. In addition to performing localization on instance search, the issue of few-shot object detection is also addressed under the same framework. Superior performance over state-of-the-art methods is observed on both tasks.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Frustratingly Simple Few-Shot Object Detection cites Meta R-CNN: Towards General Solver for Instance-Level Low-Shot Learning
  • Frustratingly Simple Few-Shot Object Detection cites Few-Shot Object Detection via Feature Reweighting
  • Frustratingly Simple Few-Shot Object Detection cites Meta-Learning to Detect Rare Objects
  • Meta R-CNN: Towards General Solver for Instance-Level Low-Shot Learning cites Few-Shot Object Detection via Feature Reweighting
  • Meta-Learning to Detect Rare Objects cites Few-Shot Object Detection via Feature Reweighting
  • Meta-RCNN: Meta Learning for Few-Shot Object Detection cites Meta R-CNN: Towards General Solver for Instance-Level Low-Shot Learning
  • Meta-RCNN: Meta Learning for Few-Shot Object Detection cites Few-Shot Object Detection via Feature Reweighting
  • Meta-RCNN: Meta Learning for Few-Shot Object Detection cites Meta-Learning to Detect Rare Objects
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on9

  1. Few-Shot Object Detection via Feature ReweightingBingyi Kang, Zhuang Liu, Xin Wang, Fisher Yu et al.ICCV 2019 · 835 citations
  2. Frustratingly Simple Few-Shot Object DetectionXin Wang, Thomas E. Huang, Joseph Gonzalez, Trevor Darrell et al.ICML 2020 · 723 citations
  3. Meta R-CNN: Towards General Solver for Instance-Level Low-Shot LearningXiaopeng Yan, Ziliang Chen, Anni Xu, Xiaoxi Wang et al.ICCV 2019 · 590 citations
  4. Meta-Learning to Detect Rare ObjectsYu-Xiong Wang, Deva Ramanan, Martial HebertICCV 2019 · 339 citations
  5. Meta-RCNN: Meta Learning for Few-Shot Object DetectionXiongwei Wu, Doyen Sahoo, Steven C. H. HoiACM MM 2020 · 94 citations

Related papers

  1. SILCO: Show a Few Images, Localize the Common ObjectTao Hu, Pascal Mettes, Jia-Hong Huang, Cees SnoekICCV 2019 · 49 citations
  2. Self-supervised object detection from audio-visual correspondenceTriantafyllos Afouras, Yuki M. Asano, Francois Fagan, Andrea Vedaldi et al.CVPR 2022 · 50 citations
  3. Rethinking the Route Towards Weakly Supervised Object LocalizationChen-Lin Zhang, Yun-Hao Cao, Jianxin WuCVPR 2020
  4. Relation-aware Instance Refinement for Weakly Supervised Visual GroundingYongfei Liu, Bo Wan, Lin Ma, Xuming HeCVPR 2021
  5. Instance Localization for Self-Supervised Detection PretrainingCeyuan Yang, Zhirong Wu, Bolei Zhou, Stephen LinCVPR 2021
Dusk over the sea between two cliffs drawn in fine vertical lines