CV Researcher, Ukrainian 🇺🇦 | PhD Student at University of Oxford, Visual Geometry Group
Joined February 2022
- Tweets106
- Following64
- Followers153
- Likes239
Orest Kupyn retweeted
Can video generation models do for vision what LLMs did for language?
Introducing GenCeption from @GoogleDeepMind: one feed-forward video model for various vision tasks — SOTA, data-efficient, and emerging behaviors (ECCV 2026)
🌐 genception.github.io
(1/8)
Orest Kupyn retweeted
Glad to see VGGRPO accepted to ECCV’26! Exciting times ahead🔥. See you all in Sweden!
🎥 Video diffusion models can create beautiful videos, but their worlds often fall apart.
🔥 Excited to share our latest work at @Google:
📢 VGGRPO: Towards World-Consistent Video Generation with 4D Latent Reward 📢
📄 arxiv.org/pdf/2603.26599
🌐 zhaochongan.github.io/projec…
🧵
📣New Paper: Feedforward Latent Triangle Splatting for Geometrically Accurate Scene Generation.
🌐 flat-splat.github.io/
🤗: huggingface.co/papers/2606.2…
A feedforward latent scene decoder for generating explicit triangle-based 3D scenes from video diffusion latents.
Another nice property is flexibility.
FLAT is trained to decode video-model latents directly, so the same scene decoder can plug into Wan-2.1 style pipelines beyond a single mode, enabling text to 3D, image to 3D, real-time generation and others.
Orest Kupyn retweeted
Excited to share our #SIGGRAPH2026 paper: Go-with-the-Track!
We utilize point-tracks as spatiotemporal conditioning to precisely insert multiple reference images into generated videos.
Paper: arxiv.org/pdf/2606.20891
Project page: eyeline-labs.github.io/Go-wi…
📢 @eyelinestudios and @netflix latest research, 🌟Go-with-the-Track🌟, has been accepted to #SIGGRAPH2026.
We propose a video generation framework that jointly conditions on multiple reference images and reference-anchored point tracks, enabling a range of content creation applications within a unified model:
• Motion-preserving video restylization using multiple reference images;
• Mesh- or keypoint-driven compositing and stylization;
• Camera motion control for both static and dynamic scenes using multiple reference images captured from different view-points and time steps.
✊Congrats to the team: @Koichi_N_, @yash2kant, Zhizheng Liu, @RyanBurgert, @Yuancheng_Xu0, @kuanhenglin, Emmett Steven, @JulienPhilip2, @maleewahaha, Andrea Vedaldi, @debfx, @realNingYu
📰 Paper: arxiv.org/pdf/2606.20891
🌐 Project: eyeline-labs.github.io/Go-wi…
⌨️ Code: github.com/Eyeline-Labs/Go-w…
*** This work is part of the ongoing research and development at @eyelinestudios and @netflix, and we look forward to seeing its techniques and workflows adopted in future productions.
Excited to release new paper: MMDiff: Extending Diffusion Transformers for Multi-Modal Generation
that was just accepted to #ECCV2026 and is finally out! 🎉
Project page 🌐: yagmurakarken.github.io/mmdi…
Paper 📄: arxiv.org/abs/2606.16673
Code: github.com/yagmurakarken/mmd…
The key idea is multi-timestep feature aggregation.
Different denoising steps capture different parts of scene structure, so reading only the final latent leaves useful information behind. Aggregating across timesteps gives much stronger representations.
Orest Kupyn retweeted
Finally the orgs, who work for Russian government and helping developing weapon in the war of aggression against Ukraine have their papers desk rejected.
Thank you, @iclr_conf.
openreview.net/forum?id=w2tn…
Orest Kupyn retweeted
VGGRPO: World-consistent video generation with 4D latent rewards
Google researchers propose a latent geometry-guided framework decoding scene geometry directly from video latents, eliminating costly VAE decoding while improving camera stability and geometric consistency in both static and dynamic scenes.
📣 New paper: PoseDreamer
A diffusion-based pipeline for generating photorealistic human data with 3D mesh supervision.
- Controllable generation + DPO
- 500k+ samples
- +76% image quality vs traditional synthetic data
- Outperforms or matches baselines across models & scales
Links:
📄 arxiv.org/pdf/2603.28763
🌐 prosperolo.github.io/posedre…
🤗 huggingface.co/datasets/pros…
Done in collaboration with @ucu_apps
Check out our recent work on geometry-aware video alignment! We enable efficient alignment by building latent geometry model that infer scene geometry directly from video latents, skipping RGB decoding entirely and making generic 4D world-consistent video alignment practical.
🎥 Video diffusion models can create beautiful videos, but their worlds often fall apart.
🔥 Excited to share our latest work at @Google:
📢 VGGRPO: Towards World-Consistent Video Generation with 4D Latent Reward 📢
📄 arxiv.org/pdf/2603.26599
🌐 zhaochongan.github.io/projec…
🧵