Leveraging 3D vision and AIGC, JD Retail’s R&D team converts abundant 2D video assets into high‑quality stereoscopic 3D space videos through a pipeline that includes monocular depth estimation, novel view synthesis, multi‑branch inpainting, and MV‑HEVC encoding, validated by ICME 2025 and a new StereoV1K dataset.
Leveraging AI-driven monocular depth estimation, novel view synthesis, and MV‑HEVC encoding, the JD Retail Content R&D team presents an end‑to‑end pipeline that converts 2D video assets into high‑quality immersive 3D spatial videos, introduces the large‑scale StereoV1K dataset, and demonstrates superior performance over existing methods.
This article presents a comprehensive AI‑driven pipeline that converts 2D video into immersive 3D spatial video by leveraging monocular depth estimation, depth‑warping novel view synthesis, a multi‑branch inpainting module, a large‑scale StereoV1K dataset, and efficient MV‑HEVC compression, with results validated at ICME 2025 and deployed in JD Vision services.
This article examines Bilibili Video Cloud’s technical exploration of spatial video and spatial audio, comparing traditional 2D encoding with MultiView (MV‑HEVC), detailing the implementation of MV‑HEVC to SBS conversion, and describing the integration of Audio Vivid technology for immersive 3‑D sound.
Apple Vision Pro supports MV‑HEVC, a multi‑view extension of HEVC that reduces inter‑view redundancy, enabling efficient 3D video encoding; Tencent’s V265 encoder implements it, achieving over 50 % bitrate savings for the secondary view and roughly 20 % overall bandwidth reduction while preserving visual quality.
This article explains the fundamentals of 3D video imaging, compares traditional 3D encoding methods with the MV-HEVC standard, and details how Tencent's V265 encoder implements MV-HEVC to achieve significant bitrate savings for Apple Vision Pro and other multiview applications.
