Detecting Humans in RGB-D Data with CNNs

Kaiyang Zhou; Adeline Paiement; Majid Mirmehdi

Detecting Humans in RGB-D Data with CNNs

Kaiyang Zhou, Adeline Paiement, Majid Mirmehdi

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

Abstract

We address the problem of people detection in RGB-D data where we leverage depth information to develop a region-of-interest (ROI) selection method that provides proposals to two color and depth CNNs. To combine the detections produced by the two CNNs, we propose a novel fusion approach based on the characteristics of depth images. We also present a new depth-encoding scheme, which not only encodes depth images into three channels but also enhances the information for classification. We conduct experiments on a publicly available RGB-D people dataset and show that our approach outperforms the baseline models that only use RGB data.

Original language	English
Title of host publication	IAPR Conference on Machine Vision Applications (MVA2017)
Publication status	Accepted/In press - 13 Feb 2017

Keywords

Digital Health

Handle.net

Persistent link

Cite this

@inproceedings{9a849d3c4d8440b58f7beca84d4ddfa1,

title = "Detecting Humans in RGB-D Data with CNNs",

abstract = "We address the problem of people detection in RGB-D data where we leverage depth information to develop a region-of-interest (ROI) selection method that provides proposals to two color and depth CNNs. To combine the detections produced by the two CNNs, we propose a novel fusion approach based on the characteristics of depth images. We also present a new depth-encoding scheme, which not only encodes depth images into three channels but also enhances the information for classification. We conduct experiments on a publicly available RGB-D people dataset and show that our approach outperforms the baseline models that only use RGB data.",

keywords = "Digital Health",

author = "Kaiyang Zhou and Adeline Paiement and Majid Mirmehdi",

year = "2017",

month = feb,

day = "13",

language = "English",

booktitle = "IAPR Conference on Machine Vision Applications (MVA2017)",

}

TY - GEN

T1 - Detecting Humans in RGB-D Data with CNNs

AU - Zhou, Kaiyang

AU - Paiement, Adeline

AU - Mirmehdi, Majid

PY - 2017/2/13

Y1 - 2017/2/13

N2 - We address the problem of people detection in RGB-D data where we leverage depth information to develop a region-of-interest (ROI) selection method that provides proposals to two color and depth CNNs. To combine the detections produced by the two CNNs, we propose a novel fusion approach based on the characteristics of depth images. We also present a new depth-encoding scheme, which not only encodes depth images into three channels but also enhances the information for classification. We conduct experiments on a publicly available RGB-D people dataset and show that our approach outperforms the baseline models that only use RGB data.

AB - We address the problem of people detection in RGB-D data where we leverage depth information to develop a region-of-interest (ROI) selection method that provides proposals to two color and depth CNNs. To combine the detections produced by the two CNNs, we propose a novel fusion approach based on the characteristics of depth images. We also present a new depth-encoding scheme, which not only encodes depth images into three channels but also enhances the information for classification. We conduct experiments on a publicly available RGB-D people dataset and show that our approach outperforms the baseline models that only use RGB data.

KW - Digital Health

M3 - Conference Contribution (Conference Proceeding)

BT - IAPR Conference on Machine Vision Applications (MVA2017)

ER -

Detecting Humans in RGB-D Data with CNNs

Abstract

Keywords

Handle.net

Fingerprint

Cite this