Jang, Deok-KyeongLee, Sung-HeeBernhard Thomaszewski and KangKang Yin and Rahul Narain2017-12-312017-12-312017978-1-4503-5091-4https://doi.org/10.1145/3099564.3106645https://diglib.eg.org:443/handle/10.1145/3099564-3106645We present a novel framework that consists of two-level regressors for finding correlations between human shapes and landmark positions in both body part and holistic scales. To this end, we first develop pose invariant coordinates of landmarks that represent both local and global shape features by using the pose invariant local shape descriptors and their spatial relationships. Our body part-level regression deals with the shape features from only those body parts corresponding to a certain landmark. For this, we develop a method that identi es such body parts per landmark, by using geometric shape dictionary obtained through the bag of features method. Our method is nearly automatic, requiring human assistance only once to di erentiate the left and right sides, and shows the prediction accuracy comparable to or better than those of existing methods, with a test data set containing a large variation of human shapes and poses.Computing methodologies Learning linear modelsShape analysisKCCAregressionsegmentationlandmark detection10.1145/3099564.3106645Deok-Kyeong Jang and Sung-Hee Lee-Computing methodologies Learning linear models; Shape analysis; KCCA, regression, segmentation, landmark detection