TM2D: Bimodality Driven 3D Dance Generation via Music-Text Integration
Published in arxiv, 2023
We propose a novel task for generating 3D dance movements that simultaneously incorporate both text and music modalities.
Publications
OmniAvatar: Geometry-Guided Controllable 3D Head Synthesis
Published in CVPR 2023, 2023
A novel geometry-guided 3D head synthesis model trained from in-the-wild unstructured images that is capable of synthesizing diverse identity-preserved 3D heads with compelling dynamic details under full disentangled control over camera poses, facial expressions, head shapes, articulated neck and jaw poses.
AgileGAN3D: Few-Shot 3D Portrait Stylization by Augmented Transfer Learning
Published in arxiv, 2023
Introduce a novel framework to produce 3D artistically appealing and personalized portraits with detailed geometry.
AvatarGen: a 3D Generative Model for Animatable Human Avatars
Published in ECCV Workshop, 2022
Introduce a novel pipeline for controllable 3D human avatar generation.
Mimicking the Oracle: An Initial Phase Decorrelation Approach for Class Incremental Learning
Published in CVPR, 2022
Introduce a novel decorrelation based method for better class incremental learning.
Volo: Vision outlooker for visual recognition
Published in arxiv, 2021
Introduce a new outlooker module and VOLO model for vision recognition tasks.
Vision permutator: A permutable mlp-like architecture for visual recognition
Published in TPAMI, 2021
Introduce a new vision permutator module and ViP model for ImageNet classification tasks.
All tokens matter: Token labeling for training better vision transformers
Published in NeurIPS, 2021
Introduce a new token labeling objective for training better vision transformers.
DeepViT: Towards Deeper Vision Transformer
Published in arxiv, 2021
Introduce a novel strategy to train deep vision transformer models and provide some insights.
Tokens-to-token vit: Training vision transformers from scratch on imagenet
Published in ICCV, 2021
Introduce T2T-ViT for ImageNet classification.
ConvBERT: Improving BERT with Span-based Dynamic Convolution
Published in NeurIPS, 2020
Introduce a novel span-based dynamic convolution module to model local attention and help improve the performance of pre-training based language models.
Reclor: A reading comprehension dataset requiring logical reasoning
Published in ICLR, 2020
Introduce a new reading comprehension dataset requiring logical reasoning.
Disentangled representation learning for 3D face shape
Published in CVPR, 2019
Introduce a disentangled representation learning framework for 3D face shape.
Joint 3d face reconstruction and dense face alignment from a single image with 2d-assisted self-supervised learning
Published in IEEE-TMM, 2019
Introduce a 3d face reconstruction framework assisted by 2d face images in the wild.