Haven (Haiwen) Feng
@HavenFeng
Followers
842
Following
2K
Statuses
120
PhD student @MPI_IS, interested in machine learning, computer vision, computer graphic, and how to understand the physical world. Visiting @berkeley_ai now.
Germany
Joined October 2021
Wow, mesh-based differentiable rendering is so back!
π’π’π’ "ππππ’ππ§π π
π¨ππ¦: Real-Time Differentiable Ray Tracing", a mesh-based 3D represention. Co-lead by my PhD students Shrisudhan Govindarajan and Daniel Rebain, and w/ @kwangmoo_yi
0
1
29
RT @ItsTheZhen: Excited to have our Nabla-GFlowNet accepted by #ICLR2025! We propose a novel GFlowNet objective to finetune a diffusion moβ¦
0
10
0
Amazing journey π! Our SGP-Bench is accepted to #ICLR2025!
π₯³Very happy that SGP-Bench is accepted to #ICLR2025! It was a wonderful collaboration with @QiuZeju, @HavenFeng, @TimZXiao, @ItsTheZhen, @katie_m_collins, Josh Tenenbaum, @adrian_weller, @Michael_J_Black and @bschoelkopf.
0
2
16
RT @davidrmcall: Decentralized Diffusion Models power stronger models trained on more accessible infrastructure. DDMs mitigate the networβ¦
0
43
0
RT @JustusThies: My group is looking for motivated PhD students that want to work on the future of digital humans. Within the ERC project 'β¦
0
16
0
RT @aomaru_21490: Introducing "AutoPresent: Designing Structured Visuals From Scratch". We employ code generation to create structured, higβ¦
0
51
0
RT @akanazawa: Did you know viser has all kinds of lights now? It's been a while, and I don't think this awesome video made by Rebecca hasβ¦
0
8
0
RT @baaadas: Nice! I wanted to say this for a long time (and I have some math for it too): DDIM sampler == rectified flow sampler, its justβ¦
0
4
0
RT @yxue_yxue: Don't miss the NeurIPS poster of Human-3Diffusion! @_R_Marin_ will present you the idea of how a 2D diffusion model and 3Dβ¦
0
6
0
RT @drjingjing2026: 1/3 Today, an anecdote shared by an invited speaker at #NeurIPS2024 left many Chinese scholars, myself included, feelinβ¦
0
629
0
RT @sunjiao123sun_: Mitigating racial bias from LLMs is a lot easier than removing it from humans! Canβt believe this happened at the besβ¦
0
904
0
RT @hhwpku: π₯ Are you ever dissatisfied with the imprecise names in vision-language datasets? π At #NeurIPS2024, we introduce ππππππππ, shβ¦
0
7
0
RT @zhengqi_li: Introducing MegaSaM! π₯ Accurate, fast, & robust structure + camera estimation from casual monocular videos of dynamic scenβ¦
0
90
0
Letβs go! But reversely this time β³
Last but not least - this morning at #ECCV2024 poster 285 "Explorative Inbetweening of Time and Space". We describe a time-reversal diffusion process that exploits a pre-trained video diffusion model to solve bounded generation tasks without fine-tuning or optimization. Create looped videos from a single image or generate the video between two frames. The approach provides insights into what video diffusion models know about the 3D world, including the movement of articulated humans.
0
0
5
Excited to present our "Time Reversal Fusion (TRF)" this morning at #ECCV2024 !!! TRF enables an image-to-video model to generate towards a given end frame without any tuning π Come and chat with us at Poster 285 during the 10:30 - 12:30 session. Video results and code available at
Thank you @_akhaliq for sharing our work! Our paper link on ArXiv: and Huggingface: BTW, the video version of our teaser figure is shown below. Explore more at:
1
6
58
Glad to help! Come to check and chat about Ruilongβs cool work this morning 10:30 - 12:30 at Poster 334 #ECCV2024
#ECCV2024 Come and checkout our poster for π NeRF-XLπ, which provides a principled solution for high fidelity city-level scale digital twins! I'm not there due to visa issues but @HavenFeng will help present this work! Cheers for Haven!
0
0
7