PanoSynthVR: Toward Light-weight 360-Degree View Synthesis from a Single Panoramic Input

Enhancing Perception and Immersion in Pre-Captured Environments through Learning-Based Eye Height Adaptation

“…However, each of these methods has its own limitations. For example, MPI representation is lightweight and can capture specular surfaces, but its discretized representation can lead to suboptimal performance for sloped surfaces [66]. In addition, MPI representation with predetermined layer structures can suffer from abrupt layer changes across discontinuities in depth, leading to inferior preserved locality.…”