Open tkuri opened 3 years ago
RGB画像のみからハンドポーズの推定を行う。まずRGBからDepth画像を推定するように回帰させる、この時UnpairedなRGB-Depthで学習させることで、データセット不足を解決。推定したDepth画像を用いてポーズ推定結果に対して正則化をかけることで性能向上するとのこと。
https://openaccess.thecvf.com/content_WACV_2020/html/Chen_DGGAN_Depth-image_Guided_Generative_Adversarial_Networks_forDisentangling_RGB_and_Depth_WACV_2020_paper.html
未確認。
論文概要
RGB画像のみからハンドポーズの推定を行う。まずRGBからDepth画像を推定するように回帰させる、この時UnpairedなRGB-Depthで学習させることで、データセット不足を解決。推定したDepth画像を用いてポーズ推定結果に対して正則化をかけることで性能向上するとのこと。
https://openaccess.thecvf.com/content_WACV_2020/html/Chen_DGGAN_Depth-image_Guided_Generative_Adversarial_Networks_forDisentangling_RGB_and_Depth_WACV_2020_paper.html
Code
未確認。