HumanRF: High-Fidelity Neural Radiance Fields for Humans in Motion

1 code implementation • 10 May 2023 • Mustafa Işık, Martin Rünz, Markos Georgopoulos, Taras Khakhulin, Jonathan Starck, Lourdes Agapito, Matthias Nießner

To close the gap to production-level quality, we introduce HumanRF, a 4D dynamic neural scene representation that captures full-body appearance in motion from multi-view video input, and enables playback from novel, unseen viewpoints.

Motion Synthesis Novel View Synthesis +1

Realistic One-shot Mesh-based Head Avatars

1 code implementation • 16 Jun 2022 • Taras Khakhulin, Vanessa Sklyarova, Victor Lempitsky, Egor Zakharov

We present a system for realistic one-shot mesh-based human head avatars creation, ROME for short.

Image Generators with Conditionally-Independent Pixel Synthesis

2 code implementations • CVPR 2021 • Ivan Anokhin, Kirill Demochkin, Taras Khakhulin, Gleb Sterkin, Victor Lempitsky, Denis Korzhenkov

Existing image generator networks rely heavily on spatial convolutions and, optionally, self-attention blocks in order to gradually synthesize images in a coarse-to-fine manner.

Image Generation

Stereo Magnification with Multi-Layer Images

2 code implementations • CVPR 2022 • Taras Khakhulin, Denis Korzhenkov, Pavel Solovev, Gleb Sterkin, Timotei Ardelean, Victor Lempitsky

The second stage infers the color and the transparency values for these layers producing the final representation for novel view synthesis.

Generalizable Novel View Synthesis Novel View Synthesis

Self-improving Multiplane-to-layer Images for Novel View Synthesis

1 code implementation • 4 Oct 2022 • Pavel Solovev, Taras Khakhulin, Denis Korzhenkov

We present a new method for lightweight novel-view synthesis that generalizes to an arbitrary forward-facing scene.

Generalizable Novel View Synthesis Novel View Synthesis

High-Resolution Daytime Translation Without Domain Labels

1 code implementation • CVPR 2020 • Ivan Anokhin, Pavel Solovev, Denis Korzhenkov, Alexey Kharlamov, Taras Khakhulin, Alexey Silvestrov, Sergey Nikolenko, Victor Lempitsky, Gleb Sterkin

We present the high-resolution daytime translation (HiDT) model for this task.

Image Manipulation Image Super-Resolution +5

DeepPavlov: Open-Source Library for Dialogue Systems

no code implementations • ACL 2018 • Mikhail Burtsev, Alex Seliverstov, er, Rafael Airapetyan, Mikhail Arkhipov, Dilyara Baymurzina, Nickolay Bushkov, Olga Gureenkova, Taras Khakhulin, Yuri Kuratov, Denis Kuznetsov, Alexey Litinsky, Varvara Logacheva, Alexey Lymar, Valentin Malykh, Maxim Petrov, Vadim Polulyakh, Leonid Pugachev, Alexey Sorokin, Maria Vikhreva, Marat Zaynutdinov

It supports modular as well as end-to-end approaches to implementation of conversational agents.

General Classification intent-classification +5

Robust Word Vectors: Context-Informed Embeddings for Noisy Texts

no code implementations • WS 2018 • Valentin Malykh, Varvara Logacheva, Taras Khakhulin

We suggest a new language-independent architecture of robust word vectors (RoVe).

Morphological Analysis Word Embeddings

Graph Convolutional Policy for Solving Tree Decomposition via Reinforcement Learning Heuristics

no code implementations • 18 Oct 2019 • Taras Khakhulin, Roman Schutski, Ivan Oseledets

We show that the agent builton GCN and trained on a single graph using an Actor-Critic method can efficiently generalize to real-world TD problem instances.

reinforcement-learning Reinforcement Learning (RL) +1

Learning Elimination Ordering for Tree Decomposition Problem

no code implementations • NeurIPS Workshop LMCA 2020 • Taras Khakhulin, Roman Schutski, Ivan Oseledets

We propose a Reinforcement Learning-based approach to approximately solve the Tree Decomposition problem.

reinforcement-learning Reinforcement Learning (RL) +1

MegaPortraits: One-shot Megapixel Neural Head Avatars

no code implementations • 15 Jul 2022 • Nikita Drobyshev, Jenya Chelishev, Taras Khakhulin, Aleksei Ivakhnenko, Victor Lempitsky, Egor Zakharov

In this work, we advance the neural head avatar technology to the megapixel resolution while focusing on the particularly challenging task of cross-driving synthesis, i. e., when the appearance of the driving image is substantially different from the animated source image.