Head pose estimation: An extensive survey on recent techniques and applications

Abate, Andrea F.; Bisogni, Carmen; Castiglione, Aniello; Nappi, Michele

doi:10.1016/j.patcog.2022.108591

Cited by 29 publications

(21 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another solution was to design a metric that was biased toward the samples with similar poses [18]. The application of HPE at RGB images is the most thriving and challenging field, since RGB images are easy to access, but sensitive to illumination and occlusion [24].…”

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

“…The input depth image was matched to the most similar synthetic models created in a training phase [26]. RGB-D images are preferred as input, especially if there exists some noise [24]. Li et al first extracted features on the RGB images, then combined depth information to obtain 3D feature points.…”

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

“…Support Vector Machine (SVM) regressors were applied to infer the head pose [28]. Although depth images can provide 3D information, its acquisition requires more sophisticated devices and controlled environment than RGB images, which limits its development in HPE [24].…”

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

“…To further improve the performance of HPE, stereo cameras are applied to obtain precise point clouds of heads. ICP is a common algorithm to compute the pose transformation between point clouds, which iteratively optimizes the transformation to minimize an error metric [24]. Tracking methods demonstrate a comparatively high level of accuracy and are not affected by identity variation.…”

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

See 3 more Smart Citations

A Deformation Field-Based Approach for Expression Processing in Head Motion Correction

2024

View full text Add to dashboard Cite

It is a routine procedure to use Iterative Closest Point (ICP) algorithm to correct motion in medical imaging of the head. However, the basic assumption of ICP is the rigid body motion, and it is not strictly valid when there is non-rigid motion (facial expression). In this study, we propose a novel method to reduce the adverse effects of facial expressions on head motion estimation. First, we design a network named DFG-Net to generate a deformation field on the RGB domain and compute a confidence map. We then extend the original ICP by embedding the confidence map representing each point's degree of non-rigid motion. Compared to other techniques, our proposed method eliminates the negative effects of facial expression while making the most of pose information. The experiments demonstrate that our approach reduces the position error by 19.23% and orientation error by 36.37% compared to the original ICP algorithm. They also show that our method is robust to variations of expressions.INDEX TERMS Non-rigid motion, ICP, deep learning, head motion correction.

show abstract

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

Section: Related Work a Head Pose Estimationmentioning

confidence: 99%

See 2 more Smart Citations

A Deformation Field-Based Approach for Expression Processing in Head Motion Correction

2024

View full text Add to dashboard Cite

show abstract

“…Extensively researched over the last 25 years [1,2], 2D head pose estimation (HPE) is a challenging but compelling and relevant computer vision problem, since several applications benefit from knowing the position of the head. This includes driving safety systems [3,4], motion capture [5], human behavior analysis in smart rooms to monitor activities [6], human interaction in meeting and workplaces [7,8], surveillance and safety [9] and target advertisement [10].…”

Section: Introductionmentioning

confidence: 99%