{"ID":61324,"CreatedAt":"2026-02-27T13:00:40Z","UpdatedAt":"2026-02-27T13:00:40Z","DeletedAt":null,"paper_url":"https://paperswithcode.com/paper/nose-eyes-and-ears-head-pose-estimation-by","arxiv_id":"1812.00739","title":"Nose, eyes and ears: Head pose estimation by locating facial keypoints","abstract":"Monocular head pose estimation requires learning a model that computes the\nintrinsic Euler angles for pose (yaw, pitch, roll) from an input image of human\nface. Annotating ground truth head pose angles for images in the wild is\ndifficult and requires ad-hoc fitting procedures (which provides only coarse\nand approximate annotations). This highlights the need for approaches which can\ntrain on data captured in controlled environment and generalize on the images\nin the wild (with varying appearance and illumination of the face). Most\npresent day deep learning approaches which learn a regression function directly\non the input images fail to do so. To this end, we propose to use a higher\nlevel representation to regress the head pose while using deep learning\narchitectures. More specifically, we use the uncertainty maps in the form of 2D\nsoft localization heatmap images over five facial keypoints, namely left ear,\nright ear, left eye, right eye and nose, and pass them through an convolutional\nneural network to regress the head-pose. We show head pose estimation results\non two challenging benchmarks BIWI and AFLW and our approach surpasses the\nstate of the art on both the datasets.","short_abstract":"Monocular head pose estimation requires learning a model that computes the intrinsic Euler angles for pose (yaw, pitch, roll) from an input image of human face.","url_abs":"http://arxiv.org/abs/1812.00739v1","url_pdf":"http://arxiv.org/pdf/1812.00739v1.pdf","authors":"[\"Aryaman Gupta\", \"Kalpit Thakkar\", \"Vineet Gandhi\", \"P. J. Narayanan\"]","published":"2018-12-03T00:00:00Z","tasks":"[\"Head Pose Estimation\", \"Pose Estimation\"]","methods":"[\"Heatmap\"]","has_code":false,"code_links":[{"ID":197241,"CreatedAt":"2026-02-27T13:01:31Z","UpdatedAt":"2026-02-27T13:01:31Z","DeletedAt":null,"paper_id":61324,"paper_url":"https://paperswithcode.com/paper/nose-eyes-and-ears-head-pose-estimation-by","paper_title":"Nose, eyes and ears: Head pose estimation by locating facial keypoints","repo_url":"https://github.com/Arnav0400/Headpose_Estimation","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"framework":"tf","github_stars":0},{"ID":497402,"CreatedAt":"2026-03-04T21:00:12Z","UpdatedAt":"2026-03-04T21:00:12Z","DeletedAt":null,"paper_id":61324,"paper_url":"https://paperswithcode.com/paper/nose-eyes-and-ears-head-pose-estimation-by","paper_title":"Nose, eyes and ears: Head pose estimation by locating facial keypoints","repo_url":"https://github.com/Arnav0400/Headpose_Estimation","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"framework":"tf","github_stars":0}]}