Wildminder (@wildmindai)
2025-12-23 | โค๏ธ 202 | ๐ 25
InfCam: Camera-controlled video generation via a depth-free homography method. Interesting, guys just freeze a Wan2.1 model & trains only lightweight attention layers. https://emjay73.github.io/InfCam/ https://x.com/wildmindai/status/2003388926787207363/video/1
๐ ์๋ณธ ๋งํฌ
๋ฏธ๋์ด
๐ Related
- neoverse-enhancing-4d-world-model-with-in-the-wild
- xie-et-al-lavr-scene-latent-conditioned-generative-video
- gamo-geometry-aware-multi-view-diffusion-outpainting-sparse
- diffusion-knows-transparency-repurposes-video-diffusion-for
- zhao-and-wei-et-al-spatia-video-generation-with-updatable