Search papers, labs, and topics across Lattice.
4
0
4
0
SBOCF slashes the number of required simulator evaluations while achieving unprecedented accuracy in estimating complex materials parameters from experimental images.
Recognizing complex, simultaneous camera movements in video clips is now feasible without the heavy computational cost of 3D models at inference.
Cross-Modal Pseudo-Labeling boosts micro-gesture recognition performance, bridging the gap between subjects and enhancing model robustness.
Current video generation benchmarks miss the forest for the trees: EvalVerse actually measures cinematic quality, not just prompt adherence.
