Automated Orientation Detection of 3D Head Reconstructions from sMRI Using Multiview Orthographic Projections: An Image Classification-Based Approach

Heredia-Lidón, Álvaro; González, Alejandro; Guerrero-Mosquera, Carlos; Gonzàlez-Colom, Rubèn; Echeverry, Luis M.; Hostalet, Noemí; Salvador, Raymond; Pomarol-Clotet, Edith; Fortea, Juan; Martínez-Abadías, Neus; Fatjó-Vilas, Mar; Sevillano, Xavier

doi:10.1007/978-3-031-36616-1_48

Cited by 1 publication

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our preliminary results indicate that the proposed approach is capable of determining the yaw angle with a very high accuracy and a fine granularity of 5 • . This validates the idea of generalising our previous proposal, which was limited to a few orthogonal orientations in the context of magnetic resonance imaging [7], to any head pose.…”

Section: Discussionsupporting

confidence: 84%

“…The main feature of our proposal involves converting the 3D head pose estimation problem into a multi-class 2D image classification problem. In that sense, each 3D head model is represented with multi-view orthographic projection images (MOPI) [7], where multiple orthogonal 2D views of the 3D head are combined into a single RGB image (see Figure 1b).…”

Section: Our Proposalmentioning

confidence: 99%

“…These methods present excellent results from 2D data. Nevertheless, there are no proposals in the literature capable of estimating full-range head orientation from 3D data, which might be useful in accurately orienting and aligning reconstructed 3D head models from structural magnetic resonance imaging (MRI) for diagnostic purposes [7].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Full-Range Yaw Prediction: A Multi-View Approach for 3D Head Model Pose Estimation Using Convolutional Neural Networks

Heredia-Lidón,

Martínez-Abadías,

Sevillano

2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Head pose estimation, a crucial task in computer vision, involves determining the orientation of a person’s head in 3D space through yaw, pitch, and roll angles. While recent techniques present excellent results in estimating head pose from a single 2D RGB image when the head faces the camera directly, few methods exist for pose estimation from arbitrary viewpoints. This problem is emphasised when the input data is in 3D, such as heads reconstructed models from magnetic resonances, where an accurate estimation of the pose is necessary for diagnostic purposes. To overcome these limitations, we make a first step by proposing a method for fine-grained head pose estimation across the full-range of yaw angles using 3D head synthetic models. Our approach involves transforming the 3D pose estimation problem into a multi-class 2D image classification problem by representing 3D head models as multi-view projection images. Leveraging a fine-tuned ResNet50 convolutional neural network, we tackle the task of head pose estimation with fine granularity of 5°, effectively discretizing the 360° yaw orientations. For the evaluation of our proposal, we train and test our models with the publicly available FaceScape and 3D BIWI datasets obtaining promising results.

show abstract

Section: Discussionsupporting

confidence: 84%

Section: Our Proposalmentioning

confidence: 99%