Search papers, labs, and topics across Lattice.
This paper introduces 4D-HOF, a feed-forward framework for 4D hand-object interaction reconstruction that bypasses costly per-sequence optimization by leveraging foundation model outputs. The core innovation is a conditional flow matching model that learns to transport initial hand-object states onto an interaction manifold, correcting pose and alignment errors in a single pass. This generative approach enables direct test-time guidance with physical constraints and 2D evidence, leading to state-of-the-art, stable, and accurate reconstructions in challenging in-the-wild scenarios.
Generative flow matching directly refines 4D hand-object interactions during synthesis, integrating physical constraints and 2D evidence without post-hoc optimization.
Existing methods for 4D hand-object reconstruction often rely on costly per-sequence optimization, while generative approaches typically synthesize interactions from random noise, which can lead to unstable interaction prediction. We introduce 4D-HOF, a feed-forward framework that reconstructs 4D hand-object interactions from coarse but informative estimates produced by vision foundation models. Concretely, we learn a conditional flow matching model that transports foundation-model-derived hand-object states toward an interaction manifold, allowing the model to correct errors in translation, rotation, and alignment in a feed-forward manner. A key advantage of our generative formulation is that it naturally enables test-time guidance within the transport process. Rather than applying a separate post-hoc optimization after reconstruction, we directly steer the evolving generative states using physical interaction constraints and observed 2D evidence, allowing the reconstruction to be refined as part of the generative process itself. By training the generative model on diverse datasets, 4D-HOF generalizes robustly to challenging in-the-wild scenarios. Experiments on out-of-domain benchmarks show that 4D-HOF achieves state-of-the-art performance, producing more stable and accurate 4D hand-object reconstructions.
