🔊 #ICCV2025 acceptance: ⚡FlashDepth⚡ estimates accurate and consistent depth for 2K-resolution videos in a real-time (24 FPS) streaming fashion on a single A100 GPU. ✊ Kudos to the teamwork led by our intern @gene_ch0u at @eyelinestudios . 👉 Join us to be the next one

Grateful to everyone who stopped by our oral presentation and posters during the whirlwind of #CVPR2025 — we know you had plenty of options!. I'm at the @eyelinestudios Booth (1209) from now–3pm today. Come say hi — I’d love to chat about our research philosophy and how it ties

🌊Go-with-the-Flow🌊, motion controllable video diffusion models, is selected as the Oral presentation at #CVPR2025 (top 0.8%). Come to watch @RyanBurgert's presentation (happening in 5min) and stop by our poster for live demos and Q&A. Our live UI is also online (done by

Attending #CVPR2025 now in #Nashville — couldn’t wait to share what we’ve been working on at @eyelinestudios , and catch up with research crowd!. ⭐ My slots @ Eyeline booth #1209 with "Go-with-the-Flow" live demos with @pablosalamancal .🗓️ June 13th Fri 2-4pm CDT.🗓️ June 15th

📚New survey out📚: “Survey of Video Diffusion Models: Foundations, Implementations, and Applications”. 85 pages, 25K words, 500 references. The so-far most comprehensive, up-to-date, and fine-grained survey milestone in this fast-evolving field, from core techniques to

If this kind of work excites you, we’re hiring @Scanline_VFX @eyelinestudios! Let’s build AI tools together for filmmaking. Research intern (open to remote, not limited to summer): Research scientist (open to remote):

The first project I led at Netflix Eyeline Studios is headed to #CVPR2025 with 5,5,4 review scores: 🌊Go-with-the-Flow🌊 warps noise for effortless motion control in video diffusion — no pipeline changes, same compute. Direct camera/object motion, transfer movement between

Excited to share our recent research 💡Lux Post Facto💡 will be presented at #CVPR2025! We’ve developed a new method to relight portrait videos in the wild with cinematic quality and temporal consistency — all in post-production. Big shoutout to @myq_1997 and the amazing team at

Thrilled to present at #CVPR2025 the first reference-based 3D-aware image 🔥Triplane Editing🔥, enabling high-quality human/animal faces editing, full-body virtual try-on, and more. Our method achieves state-of-the-art results over latent, text & image-guided 2D/3D-aware GAN &

🗓️Happening today in 11h! We are presenting 🚀T2Vs Meet VLMs🚀 at #NeurIPS2024 #vancouver West Ballroom A-D #5101, Dec 12th 11am-2pm PST. Open-source data & benchmarks drive AI innovation. We introduce a compressive image+video dataset spanning 10 harmfulness concepts, powered by

Happening today in 8h! We are presenting 🥷Shadowcast🥷 at #NeurIPS2024 #vancouver East Exhibit Hall A-C #4401, Dec 11th 11am-2pm PST. Shadowcast is the first poisoning attack against #VLMs that generates coherent but mind-bending misinformation, underscoring the necessity for

Open-source data & benchmarks drive AI innovation. We are thrilled to present 🚀T2Vs Meet VLMs🚀 at #neurips2024 D&B Track. We introduce a compressive image+video dataset spanning 10 harmfulness concepts, powered by advanced image/video diffusion models and vision-language agents

Thrilled to present 🥷Shadowcast 🥷 at #NeurIPS2024 #vancouver, the first poisoning attack against #VLMs that generates coherent but mind-bending misinformation, underscoring the necessity for responsible and trustworthy #VLM. With @Yuancheng_Xu0, Jiarui Yao, @ManliShu,

Thrilled to present 💡DifFRelight💡 at #SIGGRAPHAsia2024, a novel-view facial dynamic relighting framework based on image diffusion editing, dynamic Gaussian Splatting conditioning, integrated lighting control, and our in-house light stage hardware. Kudos to the team: Mingming