Training Object Detectors from Few Weakly-Labeled and Many Unlabeled Images

Abstract : Weakly-supervised object detection attempts to limit the amount of supervision by dispensing the need for bounding boxes, but still assumes image-level labels on the entire training set are available. In this work, we study the problem of training an object detector from one or few clean images with image-level labels and a larger set of completely unlabeled images. This is an extreme case of semi-supervised learning where the labeled data are not enough to bootstrap the learning of a classifier or detector. Our solution is to use a standard weakly-supervised pipeline to train a student model from image-level pseudo-labels generated on the unlabeled set by a teacher model, bootstrapped by region-level similarities to clean labeled images. By using the recent pipeline of PCL [47] and more unlabeled images, we achieve performance competitive or superior to many state of the art weakly-supervised detection solutions.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [61 references]  Display  Hide  Download

https://hal.inria.fr/hal-02393688
Contributor : Yannis Avrithis <>
Submitted on : Wednesday, December 4, 2019 - 2:47:01 PM
Last modification on : Friday, December 6, 2019 - 1:20:54 AM

File

1912.00384.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02393688, version 1
  • ARXIV : 1912.00384

Citation

Zhaohui Yang, Miaojing Shi, Yannis Avrithis, Chao Xu, Vittorio Ferrari. Training Object Detectors from Few Weakly-Labeled and Many Unlabeled Images. 2019. ⟨hal-02393688⟩

Share

Metrics

Record views

15

Files downloads

99