AR-1-to-3: Single Image to Consistent 3D Object Generation via Next-View Prediction
This addresses a key bottleneck in 3D content creation for applications like VR and gaming, though it is an incremental improvement over existing diffusion-based methods.
The paper tackles the problem of inconsistent novel view synthesis in image-to-3D generation by proposing AR-1-to-3, a next-view prediction method that progressively synthesizes views from close to far, resulting in significantly improved consistency and high-fidelity 3D assets.
Novel view synthesis (NVS) is a cornerstone for image-to-3d creation. However, existing works still struggle to maintain consistency between the generated views and the input views, especially when there is a significant camera pose difference, leading to poor-quality 3D geometries and textures. We attribute this issue to their treatment of all target views with equal priority according to our empirical observation that the target views closer to the input views exhibit higher fidelity. With this inspiration, we propose AR-1-to-3, a novel next-view prediction paradigm based on diffusion models that first generates views close to the input views, which are then utilized as contextual information to progressively synthesize farther views. To encode the generated view subsequences as local and global conditions for the next-view prediction, we accordingly develop a stacked local feature encoding strategy (Stacked-LE) and an LSTM-based global feature encoding strategy (LSTM-GE). Extensive experiments demonstrate that our method significantly improves the consistency between the generated views and the input views, producing high-fidelity 3D assets.