Skip to main navigation Skip to search Skip to main content

Surprisingly Easy Network Compression and Data Extension for Object Instance Detection

  • Beihang University
  • Jingchi.ai

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

To detect instances in unstructured environment with mobile system, we develop a light weight but accurate learning model denoted as B-PA(BING Pruned Alexnet). Our method first utilizes BING(Binarized Normed Gradient) to compute bounding boxes, then builds a compressed network for recognition by pruning neurons and cutting fully connected layers on the original noted Alexnet. Addressing the problem that the training samples for instance detection are limited and of small variation, we extend the training data by combining data augmentation with synthetic generation. Our B-PA model takes only 5.3MB, which is 50 times smaller but with equivalent or even higher accuracy than the original Alexnet. Experiment results demonstrate that our method outperforms the state-of-art instance detection algorithms on WRGB-D Dataset and GMU Kitchen Dataset.

Original languageEnglish
Title of host publicationVCIP 2018 - IEEE International Conference on Visual Communications and Image Processing
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538644584
DOIs
StatePublished - 2 Jul 2018
Event33rd IEEE International Conference on Visual Communications and Image Processing, VCIP 2018 - Taichung, Taiwan, Province of China
Duration: 9 Dec 201812 Dec 2018

Publication series

NameVCIP 2018 - IEEE International Conference on Visual Communications and Image Processing

Conference

Conference33rd IEEE International Conference on Visual Communications and Image Processing, VCIP 2018
Country/TerritoryTaiwan, Province of China
CityTaichung
Period9/12/1812/12/18

Keywords

  • Bi-narised Normed Gradient
  • Data Extension; Synthetic generation
  • Object Instance Detection
  • Pruned Alexnet

Fingerprint

Dive into the research topics of 'Surprisingly Easy Network Compression and Data Extension for Object Instance Detection'. Together they form a unique fingerprint.

Cite this