AD-NeRF: Audio Driven Neural Radiance Fields for Talking Head Synthesis pdf: arxiv.org/pdf/2103.11078… abs: arxiv.org/abs/2103.11078

"Delving Deep into the Generalization of Vision Transformers under Distribution Shifts": Paper: arxiv.org/pdf/2106.07617… Code: github.com/Phoenix1153/Vi… - We investigate the OOD generalization of vision transformers. - We integrated domain adaptation techniques into transformers.

Interested in knowing more about our #CVPR2021 paper on #FaceVid2Vid, we welcome you to test out the algorithm yourself. Our online demo is available at nvidia.com/en-us/research…

A Survey on Neural Speech Synthesis pdf: arxiv.org/pdf/2106.15561… abs: arxiv.org/abs/2106.15561

Real-Time Super-Resolution System of 4K-Video Based on Deep Learning pdf: arxiv.org/pdf/2107.05307… abs: arxiv.org/abs/2107.05307 github: github.com/Thmen/EGVSR

Wordcraft: a Human-AI Collaborative Editor for Story Writing pdf: arxiv.org/pdf/2107.07430… abs: arxiv.org/abs/2107.07430 an AI-assisted editor for story writing in which a writer and a dialog system collaborate to write a story

StyleVideoGAN: A Temporal Generative Model using a Pretrained StyleGAN pdf: arxiv.org/pdf/2107.07224… abs: arxiv.org/abs/2107.07224 a temporal GAN for the unconditional generation of high-quality videos

How to do research with my mentors effectively? I get this question frequently in my open office hours. I am still learning as well but I hope sharing my ✌💰 may be helpful to some. Key idea ➡️ **Help them help you!** How? Check out the thread 🧵

How to make steady progress in my research? I worked so damn hard but "IT JUST DOESN'T WORK!"😤 How can I unblock myself quickly and make good progress toward the goals? Below I compiled a list of tips that I found useful. 👇

Thrilled to share my very first SIGGRAPH Asia paper! "Pose with Style". We generate photorealistic detail-preserving human reposing results and virtual try-on. Project: pose-with-style.github.io

Controllable and Compositional Generation with Latent-Space Energy-Based Models abs: arxiv.org/abs/2110.10873 by composing energy functions with logical operators, this work is the first to achieve such compositionality in generating photo-realistic images of resolution 1024×1024

PatchFormer: A Versatile 3D Transformer Based on Patch Attention abs: arxiv.org/abs/2111.00207

Unsupervised Learning of Compositional Energy Concepts abs: arxiv.org/abs/2111.03042 project page: energy-based-model.github.io/comet/

Our #NeurIPS work on Panoptic 3D Scene Reconstruction from a Single RGB Image! We first predict 2D features, lift these to 3D, and jointly predict geometry, semantic and instances. Work by Manuel Dahnert Vuthithang Angela Dai arxiv.org/abs/2111.02444 youtu.be/YVxRNHmd5SA

GRAM: Generative Radiance Manifolds for 3D-Aware Image Generation abs: arxiv.org/abs/2112.08867 project page: yudeng.github.io/GRAM/

StyleGAN-V: A Continuous Video Generator with the Price, Image Quality and Perks of StyleGAN2 abs: arxiv.org/abs/2112.14683 project page: universome.github.io/stylegan-v can generate arbitrarily long videos at arbitrary high frame rate

To help build more versatile & robust AI speech recognition tools, we are announcing Audio-Visual HuBERT (AV-HuBERT), a state-of-the-art self-supervised framework for understanding speech that learns by observing & hearing people speak. ai.facebook.com/blog/ai-that-u…

Learning with Neighbor Consistency for Noisy Labels abs: arxiv.org/abs/2202.02200

Want to extract triangle meshes, materials, and lighting directly from images? Paper, video & code: nvlabs.github.io/nvdiffrec/ CVPR 2022 (Oral). Nice collaboration between Aaron Lefohn and Sanja Fidler teams at NVIDIA.

Machine Learning Operations (MLOps): Overview, Definition, and Architecture arxiv.org/abs/2205.02302