no code implementations • 21 May 2024 • Jia Gong, Shenyu Ji, Lin Geng Foo, Kang Chen, Hossein Rahmani, Jun Liu
To generate high-quality garments for each layer, we introduce a coarse-to-fine strategy for diverse garment generation and a novel dual-SDS loss function to maintain coherence between the generated garments and avatar components, including the human body and other garments.
no code implementations • 1 Apr 2024 • Lin Geng Foo, Tianjiao Li, Hossein Rahmani, Jun Liu
Action detection aims to localize the starting and ending points of action instances in untrimmed videos, and predict the classes of those instances.
no code implementations • 1 Apr 2024 • Jia Gong, Lin Geng Foo, Yixuan He, Hossein Rahmani, Jun Liu
Sign Language Translation (SLT) is a challenging task that aims to translate sign videos into spoken language.
Ranked #1 on Gloss-free Sign Language Translation on PHOENIX14T
Gloss-free Sign Language Translation Sign Language Translation +1
no code implementations • 27 Aug 2023 • Lin Geng Foo, Hossein Rahmani, Jun Liu
Due to its wide range of applications and the demonstrated potential of recent works, AIGC developments have been attracting lots of attention recently, and AIGC methods have been developed for various data modalities, such as image, video, text, 3D shape (as voxels, point clouds, meshes, and neural implicit fields), 3D scene, 3D human avatar (body and head), 3D motion, and audio -- each presenting different characteristics and challenges.
no code implementations • ICCV 2023 • Lin Geng Foo, Jia Gong, Hossein Rahmani, Jun Liu
Inspired by their capability, we explore a diffusion-based approach for human mesh recovery, and propose a Human Mesh Diffusion (HMDiff) framework which frames mesh recovery as a reverse diffusion process.
no code implementations • CVPR 2023 • Tianjiao Li, Lin Geng Foo, Ping Hu, Xindi Shang, Hossein Rahmani, Zehuan Yuan, Jun Liu
Pre-training VTs on such corrupted data can be challenging, especially when we pre-train via the masked autoencoding approach, where both the inputs and masked ``ground truth" targets can potentially be unreliable in this case.
no code implementations • 1 Apr 2023 • Jianhong Pan, Siyuan Yang, Lin Geng Foo, Qiuhong Ke, Hossein Rahmani, Zhipeng Fan, Jun Liu
Currently, salience-based channel pruning makes continuous breakthroughs in network compression.
no code implementations • 1 Apr 2023 • Jianhong Pan, Lin Geng Foo, Qichen Zheng, Zhipeng Fan, Hossein Rahmani, Qiuhong Ke, Jun Liu
Dynamic neural networks can greatly reduce computation redundancy without compromising accuracy by adapting their structures based on the input.
no code implementations • CVPR 2023 • Lin Geng Foo, Jia Gong, Zhipeng Fan, Jun Liu
Recent years have witnessed great progress in deep neural networks for real-time applications.
no code implementations • CVPR 2023 • Lin Geng Foo, Tianjiao Li, Hossein Rahmani, Qiuhong Ke, Jun Liu
We propose a Unified Pose Sequence Modeling approach to unify heterogeneous human behavior understanding tasks based on pose data, e. g., action recognition, 3D pose estimation and 3D early action prediction.
no code implementations • CVPR 2023 • Haoxuan Qu, Yujun Cai, Lin Geng Foo, Ajay Kumar, Jun Liu
Therefore, via minimizing the distance between the two characteristic functions, we can optimize the model to provide a more accurate localization result for the body joints in different sub-regions of the predicted heatmap.
1 code implementation • CVPR 2023 • Jia Gong, Lin Geng Foo, Zhipeng Fan, Qiuhong Ke, Hossein Rahmani, Jun Liu
Monocular 3D human pose estimation is quite challenging due to the inherent ambiguity and occlusion, which often lead to high uncertainty and indeterminacy.
Ranked #11 on 3D Human Pose Estimation on MPI-INF-3DHP
no code implementations • 13 Oct 2022 • Haoxuan Qu, Yanchao Li, Lin Geng Foo, Jason Kuen, Jiuxiang Gu, Jun Liu
Confidence estimation, a task that aims to evaluate the trustworthiness of the model's prediction output during deployment, has received lots of research attention recently, due to its importance for the safe deployment of deep models.
no code implementations • 3 Oct 2022 • Haoxuan Qu, Li Xu, Yujun Cai, Lin Geng Foo, Jun Liu
In this paper, we show that optimizing the heatmap prediction in such a way, the model performance of body joint localization, which is the intrinsic objective of this task, may not be consistently improved during the optimization process of the heatmap prediction.
no code implementations • 3 Sep 2022 • Tianjiao Li, Lin Geng Foo, Qiuhong Ke, Hossein Rahmani, Anran Wang, Jinghua Wang, Jun Liu
We design a novel Dynamic Spatio-Temporal Specialization (DSTS) module, which consists of specialized neurons that are only activated for a subset of samples that are highly similar.
no code implementations • 20 Jul 2022 • Lin Geng Foo, Tianjiao Li, Hossein Rahmani, Qiuhong Ke, Jun Liu
Early action prediction aims to successfully predict the class label of an action before it is completely performed.
no code implementations • 21 Jul 2020 • Lin Geng Foo, Rui En Ho, Jiamei Sun, Alexander Binder
In this work, we propose a two-step post-processing procedure, Split and Expand, that directly improves the conversion of segmentation maps to instances.