Search papers, labs, and topics across Lattice.
KAIST
6
0
5
GeoCore-9B sets a new benchmark in Earth observation generative modeling by achieving unprecedented accuracy and versatility through geospatially aware training.
AVSR-Diff achieves unprecedented temporal consistency and detail preservation in arbitrary-scale video super-resolution, outperforming traditional fixed-scale models.
WebSpline achieves unprecedented rendering speed and quality in dynamic 3D scene reconstruction from monocular videos, outperforming existing techniques by a significant margin.
Finally, a diffusion model lets you puppeteer multiple objects in a video with just text prompts, opening the door to complex scene editing.
Rendering artifacts in feed-forward 3D Gaussian Splatting? Solved: AA-Splat delivers a whopping 7dB PSNR boost by fixing screen-space dilation filters.
Skip the expensive paired video data: PropFly trains video editing propagation models by generating supervision signals directly from pre-trained video diffusion models.
