RGB-D Image Understanding using Supervision Transfer

Saurabh Gupta, Judy Hoffman, Jitendra Malik

Research output: Contribution to journalConference articlepeer-review

Abstract

Current advances in object recognition and scene understanding have been enabled by the availability of large number of labeled images. Similar advances in RGB-D image understanding are hampered by the current lack of large labeled datasets in this domain. We have a developed a new technique “cross-modal distillation” which enables us to transfer supervision from RGB to RGB-D datasets. We use representations learned from labeled RGB images as a supervisory signal to train representations for depth images, and observe a 6% relative gain in performance for object detection with RGB-D images, and a 20% relative improvement when only using the depth image.

Original languageEnglish (US)
Pages (from-to)444-447
Number of pages4
JournalDigest of Technical Papers - SID International Symposium
Volume47
Issue number1
DOIs
StatePublished - 2016
Externally publishedYes
Event54th Annual SID Symposium, Seminar, and Exhibition 2016, Display Week 2016 - San Francisco, United States
Duration: May 22 2016May 27 2016

Keywords

  • RGB-D Scene Understanding

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'RGB-D Image Understanding using Supervision Transfer'. Together they form a unique fingerprint.

Cite this