Simplified Regression for Human Pose Estimation
碩士 === 國立清華大學 === 資訊工程學系 === 104 === We present a two-stage deep convolutional neural network for human pose estimation. In the first stage, it directly extracts features from the input image and combines all the features to generate a compact yet effective result for predicting the keypoint locatio...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2016
|
Online Access: | http://ndltd.ncl.edu.tw/handle/76196911503341670670 |
id |
ndltd-TW-104NTHU5392088 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-104NTHU53920882017-08-27T04:30:16Z http://ndltd.ncl.edu.tw/handle/76196911503341670670 Simplified Regression for Human Pose Estimation 用於人體姿勢估測之簡化回歸方法 Chen, Yu Fu 陳昱甫 碩士 國立清華大學 資訊工程學系 104 We present a two-stage deep convolutional neural network for human pose estimation. In the first stage, it directly extracts features from the input image and combines all the features to generate a compact yet effective result for predicting the keypoint locations instead of producing one heatmap for each keypoint. Then, we use the input image and the synthetic heatmaps derived from the previous stage as the input of the second stage to get a refined result of pose estimation. We evaluate our method on two datasets: FLIC and LSP. Our method achieves the state-of-the-art performance on FLIC dataset. Chen, Hwann Tzong 陳煥宗 2016 學位論文 ; thesis 27 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立清華大學 === 資訊工程學系 === 104 === We present a two-stage deep convolutional neural network for human pose estimation. In the first stage, it directly extracts features from the input image and combines all the features to generate a compact yet effective result for predicting the keypoint locations instead of producing one heatmap for each keypoint. Then, we use the input image and the synthetic heatmaps derived from the previous stage as the input of the second stage to get a refined result of pose estimation. We evaluate our method on two datasets: FLIC and LSP. Our method achieves the state-of-the-art performance on FLIC dataset.
|
author2 |
Chen, Hwann Tzong |
author_facet |
Chen, Hwann Tzong Chen, Yu Fu 陳昱甫 |
author |
Chen, Yu Fu 陳昱甫 |
spellingShingle |
Chen, Yu Fu 陳昱甫 Simplified Regression for Human Pose Estimation |
author_sort |
Chen, Yu Fu |
title |
Simplified Regression for Human Pose Estimation |
title_short |
Simplified Regression for Human Pose Estimation |
title_full |
Simplified Regression for Human Pose Estimation |
title_fullStr |
Simplified Regression for Human Pose Estimation |
title_full_unstemmed |
Simplified Regression for Human Pose Estimation |
title_sort |
simplified regression for human pose estimation |
publishDate |
2016 |
url |
http://ndltd.ncl.edu.tw/handle/76196911503341670670 |
work_keys_str_mv |
AT chenyufu simplifiedregressionforhumanposeestimation AT chényùfǔ simplifiedregressionforhumanposeestimation AT chenyufu yòngyúréntǐzīshìgūcèzhījiǎnhuàhuíguīfāngfǎ AT chényùfǔ yòngyúréntǐzīshìgūcèzhījiǎnhuàhuíguīfāngfǎ |
_version_ |
1718519378890194944 |