Energy flow: image correspondence approximation for motion analysis
We propose a correspondence approximation approach between temporally adjacent frames for motion analysis. First, energy map is established to represent image spatial features on multiple scales using Gaussian convolution. On this basis, energy flow at each layer is estimated using Gauss–Seidel iter...
Main Authors: | , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | en_US |
Published: |
SPIE
2016
|
Online Access: | http://hdl.handle.net/1721.1/103903 |
_version_ | 1811068722522095616 |
---|---|
author | Wang, Liangliang Li, Ruifeng Fang, Yajun |
author2 | Massachusetts Institute of Technology. Microsystems Technology Laboratories |
author_facet | Massachusetts Institute of Technology. Microsystems Technology Laboratories Wang, Liangliang Li, Ruifeng Fang, Yajun |
author_sort | Wang, Liangliang |
collection | MIT |
description | We propose a correspondence approximation approach between temporally adjacent frames for motion analysis. First, energy map is established to represent image spatial features on multiple scales using Gaussian convolution. On this basis, energy flow at each layer is estimated using Gauss–Seidel iteration according to the energy invariance constraint. More specifically, at the core of energy invariance constraint is “energy conservation law” assuming that the spatial energy distribution of an image does not change significantly with time. Finally, energy flow field at different layers is reconstructed by considering different smoothness degrees. Due to the multiresolution origin and energy-based implementation, our algorithm is able to quickly address correspondence searching issues in spite of background noise or illumination variation. We apply our correspondence approximation method to motion analysis, and experimental results demonstrate its applicability. |
first_indexed | 2024-09-23T08:00:08Z |
format | Article |
id | mit-1721.1/103903 |
institution | Massachusetts Institute of Technology |
language | en_US |
last_indexed | 2024-09-23T08:00:08Z |
publishDate | 2016 |
publisher | SPIE |
record_format | dspace |
spelling | mit-1721.1/1039032022-09-23T10:14:26Z Energy flow: image correspondence approximation for motion analysis Wang, Liangliang Li, Ruifeng Fang, Yajun Massachusetts Institute of Technology. Microsystems Technology Laboratories Fang, Yajun We propose a correspondence approximation approach between temporally adjacent frames for motion analysis. First, energy map is established to represent image spatial features on multiple scales using Gaussian convolution. On this basis, energy flow at each layer is estimated using Gauss–Seidel iteration according to the energy invariance constraint. More specifically, at the core of energy invariance constraint is “energy conservation law” assuming that the spatial energy distribution of an image does not change significantly with time. Finally, energy flow field at different layers is reconstructed by considering different smoothness degrees. Due to the multiresolution origin and energy-based implementation, our algorithm is able to quickly address correspondence searching issues in spite of background noise or illumination variation. We apply our correspondence approximation method to motion analysis, and experimental results demonstrate its applicability. National Natural Science Foundation (China) (Grant No. 661273339) 2016-08-11T19:22:39Z 2016-08-11T19:22:39Z 2016-04 2015-12 Article http://purl.org/eprint/type/JournalArticle 0091-3286 http://hdl.handle.net/1721.1/103903 Wang, Liangliang, Ruifeng Li, and Yajun Fang. "Energy flow: image correspondence approximation for motion analysis." Optical Engineering 55:4 (April 2016), 043109-1. en_US http://dx.doi.org/10.1117/1.oe.55.4.043109 Optical Engineering Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf SPIE SPIE |
spellingShingle | Wang, Liangliang Li, Ruifeng Fang, Yajun Energy flow: image correspondence approximation for motion analysis |
title | Energy flow: image correspondence approximation for motion analysis |
title_full | Energy flow: image correspondence approximation for motion analysis |
title_fullStr | Energy flow: image correspondence approximation for motion analysis |
title_full_unstemmed | Energy flow: image correspondence approximation for motion analysis |
title_short | Energy flow: image correspondence approximation for motion analysis |
title_sort | energy flow image correspondence approximation for motion analysis |
url | http://hdl.handle.net/1721.1/103903 |
work_keys_str_mv | AT wangliangliang energyflowimagecorrespondenceapproximationformotionanalysis AT liruifeng energyflowimagecorrespondenceapproximationformotionanalysis AT fangyajun energyflowimagecorrespondenceapproximationformotionanalysis |