arXiv reaDer
EEG-Driven 3D Object Reconstruction with Color Consistency and Diffusion Prior
EEG-based visual perception reconstruction has become a current research hotspot. Neuroscientific studies have shown that humans can perceive various types of visual information, such as color, shape, and texture, when observing objects. However, existing technical methods often face issues such as inconsistencies in texture, shape, and color between the visual stimulus images and the reconstructed images. In this paper, we propose a method for reconstructing 3D objects with color consistency based on EEG signals. The method adopts a two-stage strategy: in the first stage, we train an implicit neural EEG encoder with the capability of perceiving 3D objects, enabling it to capture regional semantic features; in the second stage, based on the latent EEG codes obtained in the first stage, we integrate a diffusion model, neural style loss, and NeRF to implicitly decode the 3D objects. Finally, through experimental validation, we demonstrate that our method can reconstruct 3D objects with color consistency using EEG.
updated: Tue Oct 29 2024 13:11:03 GMT+0000 (UTC)
published: Mon Oct 28 2024 12:59:24 GMT+0000 (UTC)
参考文献 (このサイトで利用可能なもの) / References (only if available on this site)
被参照文献 (このサイトで利用可能なものを新しい順に) / Citations (only if available on this site, in order of most recent)
Amazon.co.jpアソシエイト