Embedding Novel Views in a Single JPEG Image
This addresses the problem of compactly storing 3D scene data for applications like virtual reality, though it is incremental as it builds on existing multiplane image representations.
The paper tackles embedding novel views into a single JPEG image, achieving high-fidelity restoration with an embedding capacity of 1024 bits per pixel and robustness to compression and modifications.
We propose a novel approach for embedding novel views in a single JPEG image while preserving the perceptual fidelity of the modified JPEG image and the restored novel views. We adopt the popular novel view synthesis representation of multiplane images (MPIs). Our model first encodes 32 MPI layers (totally 128 channels) into a 3-channel JPEG image that can be decoded for MPIs to render novel views, with an embedding capacity of 1024 bits per pixel. We conducted experiments on public datasets with different novel view synthesis methods, and the results show that the proposed method can restore high-fidelity novel views from a slightly modified JPEG image. Furthermore, our method is robust to JPEG compression, color adjusting, and cropping. Our source code will be publicly available.