tttLRM: Test-Time Training for Long Context and Autoregressive 3D Reconstruction
This addresses the challenge of efficient and high-quality 3D reconstruction for computer vision applications, with incremental improvements in scaling and performance.
The paper tackles the problem of 3D reconstruction from multiple images by proposing tttLRM, a model that uses test-time training to enable long-context, autoregressive reconstruction with linear complexity, achieving superior performance in feedforward 3D Gaussian reconstruction compared to state-of-the-art methods on objects and scenes.
We propose tttLRM, a novel large 3D reconstruction model that leverages a Test-Time Training (TTT) layer to enable long-context, autoregressive 3D reconstruction with linear computational complexity, further scaling the model's capability. Our framework efficiently compresses multiple image observations into the fast weights of the TTT layer, forming an implicit 3D representation in the latent space that can be decoded into various explicit formats, such as Gaussian Splats (GS) for downstream applications. The online learning variant of our model supports progressive 3D reconstruction and refinement from streaming observations. We demonstrate that pretraining on novel view synthesis tasks effectively transfers to explicit 3D modeling, resulting in improved reconstruction quality and faster convergence. Extensive experiments show that our method achieves superior performance in feedforward 3D Gaussian reconstruction compared to state-of-the-art approaches on both objects and scenes.