Selective Search

Selective Search is a region proposal algorithm for object detection tasks. It starts by over-segmenting the image based on intensity of the pixels using a graph-based segmentation method by Felzenszwalb and Huttenlocher. Selective Search then takes these oversegments as initial input and performs the following steps

  1. Add all bounding boxes corresponding to segmented parts to the list of regional proposals
  2. Group adjacent segments based on similarity
  3. Go to step 1

At each iteration, larger segments are formed and added to the list of region proposals. Hence we create region proposals from smaller segments to larger segments in a bottom-up approach. This is what we mean by computing “hierarchical” segmentations using Felzenszwalb and Huttenlocher’s oversegments.

Latest Papers

PAPER DATE
Learning Objectness from Sonar Images for Class-Independent Object Detection
Matias Valdenegro-Toro
2019-07-01
You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection
Krishna Kumar Singh Yong Jae Lee
2019-06-01
RRPN: Radar Region Proposal Network for Object Detection in Autonomous Vehicles
| Ramin NabatiHairong Qi
2019-05-01
Automatic Handgun Detection in X-ray Images using Bag of Words Model with Selective Search
David Castro PiñolEnrique Juan Marañón Reyes
2019-03-04
Semantic Hierarchical Priors for Intrinsic Image Decomposition
Saurabh SainiP. J. Narayanan
2019-02-11
Learning Position Evaluation Functions Used in Monte Carlo Softmax Search
Harukazu IgarashiYuichi MoriokaKazumasa Yamamoto
2019-01-30
Deep Multiple Instance Learning for Zero-shot Image Tagging
Shafin RahmanSalman Khan
2018-03-16
ME R-CNN: Multi-Expert R-CNN for Object Detection
Hyungtae LeeSungmin EumHeesung Kwon
2017-04-04
Deep Learning the Indus Script
Satish PalaniappanRonojoy Adhikari
2017-02-02
A MultiPath Network for Object Detection
| Sergey ZagoruykoAdam LererTsung-Yi LinPedro O. PinheiroSam GrossSoumith ChintalaPiotr Dollár
2016-04-07
Diversity in Object Proposals
Anton WinschelRainer LienhartChristian Eggert
2016-03-14
Evaluation of Object Detection Proposals Under Condition Variations
Fahimeh RezazadeganSareh ShiraziMichael MilfordBen Upcroft
2015-12-10
Unsupervised Tube Extraction Using Transductive Learning and Dense Trajectories
Mihai Marian PuscasEnver SanginetoDubravko CulibrkNicu Sebe
2015-12-01
Boosting Convolutional Features for Robust Object Proposals
Nikolaos KarianakisThomas J. FuchsStefano Soatto
2015-03-21

Components

COMPONENT TYPE
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign

Categories