Abstract:
This paper introduces a framework to recover avatar details from a single image. Human shapes, body poses, textures, and viewpoints make it difficult. Previous methods used a parametric template without surface details to recover the human body shape. Thus, the body appears naked. We propose a new learning-based framework that combines parametric model robustness with free-form 3D deformation. In a Hierarchical Mesh Deformation (HMD) framework, body joints, silhouettes, and per-pixel shading information constrain deep neural networks to refine the 3D shape. Our method restores detailed human body shapes and textures beyond skinned models. Our method outperforms state-of-the-art methods in 2D IoU number and 3D metric distance accuracy.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
