Understanding the effects of pre-training for object detectors via eigenspectrum

Yosuke Shinya, Edgar Simo-Serra, Taiji Suzuki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

ImageNet pre-training has been regarded as essential for training accurate object detectors for a long time. Recently, it has been shown that object detectors trained from randomly initialized weights can be on par with those fine-tuned from ImageNet pre-trained models. However, the effects of pre-training and the differences caused by pre-training are still not fully understood. In this paper, we analyze the eigenspectrum dynamics of the covariance matrix of each feature map in object detectors. Based on our analysis on ResNet-50, Faster R-CNN with FPN, and Mask R-CNN, we show that object detectors trained from ImageNet pre-trained models and those trained from scratch behave differently from each other even if both object detectors have similar accuracy. Furthermore, we propose a method for automatically determining the widths (the numbers of channels) of object detectors based on the eigenspectrum. We train Faster R-CNN with FPN from randomly initialized weights, and show that our method can reduce ~27% of the parameters of ResNet-50 without increasing Multiply-Accumulate operations and losing accuracy. Our results indicate that we should develop more appropriate methods for transferring knowledge from image classification to object detection (or other tasks).

Original languageEnglish
Title of host publicationProceedings - 2019 International Conference on Computer Vision Workshop, ICCVW 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1931-1941
Number of pages11
ISBN (Electronic)9781728150239
DOIs
Publication statusPublished - 2019 Oct
Event17th IEEE/CVF International Conference on Computer Vision Workshop, ICCVW 2019 - Seoul, Korea, Republic of
Duration: 2019 Oct 272019 Oct 28

Publication series

NameProceedings - 2019 International Conference on Computer Vision Workshop, ICCVW 2019

Conference

Conference17th IEEE/CVF International Conference on Computer Vision Workshop, ICCVW 2019
CountryKorea, Republic of
CitySeoul
Period19/10/2719/10/28

Keywords

  • Eigenspectrum
  • Fine tuning
  • Intrinsic architecture
  • Knowledge transfer
  • Neural architecture search
  • Object detection
  • Pre training

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Fingerprint Dive into the research topics of 'Understanding the effects of pre-training for object detectors via eigenspectrum'. Together they form a unique fingerprint.

  • Cite this

    Shinya, Y., Simo-Serra, E., & Suzuki, T. (2019). Understanding the effects of pre-training for object detectors via eigenspectrum. In Proceedings - 2019 International Conference on Computer Vision Workshop, ICCVW 2019 (pp. 1931-1941). [9022267] (Proceedings - 2019 International Conference on Computer Vision Workshop, ICCVW 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCVW.2019.00242