Jiaheng Liu

Jiaheng Liu

Auto-generated profile, based on publicly available information

Jiaheng Liu

Nanjing University, Nanjing, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2024
Verified
Conference Paper
Open access
PDF file
NWNoah Wang
Noah Wang
ZPZ.Y. Peng
Z.Y. Peng
HQHaoran Que
Haoran Que
JLJiaheng Liu
Jiaheng Liu
WZWangchunshu Zhou
Wangchunshu Zhou
JPJunran Peng
Junran Peng
Noah Wang, Z.y. Peng, Haoran Que, Jiaheng Liu, Wangchunshu Zhou, Yuhan Wu, Hongcheng Guo, Ruitong Gan, Zehao Ni, Jian Yang, Man Zhang, Zhaoxiang Zhang, Wanli Ouyang, Ke Xu, Wenhao Huang, Jie Fu, Junran Peng. Findings of the Association for Computational Linguistics ACL 2024. 2024.

1 January 2021 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
JLJiaheng Liu
Jiaheng Liu
SZShunfeng Zhou
Shunfeng Zhou
YWYichao Wu
Yichao Wu
KCKen Chen
Ken Chen
WOWanli Ouyang
Wanli Ouyang
DXDong Xu
Dong Xu
Typical neural architecture search methods usually restrict the search space to the pre-defined block types for a fixed macro-architecture. However, this strategy will affect the search space and architecture flexibility if block proposal search (BPS) is not considered for NAS. As a result, block structure search is the bottleneck in many previous NAS works. In this paper, we propose a new evolutionary algorithm referred as latency EvoNAS (LEvoNAS) for block structure search, and also introduce it to the NAS framework by developing a novel two-stage framework called Block Proposal NAS (BP-NAS). Comprehensive experimental results across multiple computer vision tasks demonstrate the superiority of our approaches over the state-of-the-art lightweight models. For the classification task on the ImageNet dataset, our BPN-A is better than 1.0-MobileNetV2 [1] with similar latency, and our BPN-B saves 23.7% latency when compared with 1.4-MobileNetV2 with higher top-1 accuracy. Furthermore, for the object detection task on the COCO dataset, the face identification task on the MegaFace dataset, and the re-identification task on Market-1501 dataset, our methods outperform MobileNetV2, which demonstrates the generalization capability of our newly proposed framework.

2 August 2021 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
JLJiaheng Liu
Jiaheng Liu
DXDong Xu
Dong Xu
In this work, we propose a strong two-stream baseline method referred to as GeometryMotion-Net for 3D action recognition. For efficient 3D action recognition, we first represent each point cloud sequence as a limited number of randomly sampled frames with each frame consisting of a sparse set of points. After that, we propose a new two-stream framework for effective 3D action recognition. For the geometry stream, we propose a new module to produce a virtual overall geometry point cloud by first merging all 3D points from these selected frames, and then we exploit local neighborhood information of each point in the feature space. In the motion stream, for any two neighboring point cloud frames, we also propose a new module to generate one virtual forward motion point cloud and one virtual backward motion point cloud. Specifically, for each point in the current frame, we first produce a set of 3D offset features relative to the neighboring points in the reference frame ( i.e. , the previous/subsequent frame) and then exploit local neighborhood information of this point in the offset feature space. Based on the newly generated virtual overall geometry point cloud and multiple virtual forward/backward motion point clouds, any existing point cloud analysis methods ( e.g. , PointNet) can be readily adopted for extracting discriminant geometry and bidirectional motion features in the geometry and motion streams, respectively, which are further aggregated to make our two-stream network trainable in an end-to-end fashion. Comprehensive experiments on both large-scale datasets ( i.e. NTU RGB+D 60 and NTU RGB+D 120) and small-scale datasets ( i.e. , N-UCLA and UWA3DII) demonstrate the effectiveness and efficiency of our two-stream network for 3D action recognition.

18 August 2021 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
Jinyang GuoJinyang Guo
Jinyang Guo
JLJiaheng Liu
Jiaheng Liu
DXDong Xu
Dong Xu
Deep neural networks designed for point clouds, also called point cloud neural networks (PCNNs), are attracting increasing attention in recent years. In this work, we propose the first model compression framework referred to as JointPruning (JP) that is specifically designed for compressing PCNNs. Observing that the redundancies in PCNNs are largely affected by certain parameters like the number of points, we first propose a new search space specifically designed for PCNNs. By searching the optimal pruning policy in our newly proposed search space, our JP framework can prune the PCNNs at different levels and simultaneously reduce the redundancies along multiple dimensions. As the newly proposed search space consists of multiple levels and the policy value at each level is continuous in our JP framework, it is hard to directly search for the best pruning policy in such a large search space. To this end, we further propose two strategies called search space refinement and validation set extension to progressively refine the granularity of our searching process in a coarse-to-fine and easy-to-hard fashion, which can help us gradually find better pruning policies. Comprehensive experiments on three benchmark datasets demonstrate the effectiveness of our JP framework for compressing PCNNs.

10 August 2022 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
JLJiaheng Liu
Jiaheng Liu
Jinyang GuoJinyang Guo
Jinyang Guo
DXDong Xu
Dong Xu
In this work, we propose a new end-to-end optimized two-stream framework called GeometryMotion-Transformer (GMT) for 3D action recognition. We first observe that the existing 3D action recognition approaches cannot well extract motion representations from point cloud sequences. Specifically, when extracting motion representations, the existing approaches do not explicitly consider one-to-one correspondence among frames. Besides, the existing methods only extract the single-scale motion representations, which cannot well model the complex motion patterns of moving objects in point cloud sequences. To address these issues, we first propose the feature extraction module (FEM) to generate one-to-one correspondence among frames without using the voxelization process, and explicitly extract both geometry and multi-scale motion representations from raw point clouds. Moreover, we also observe the existing two-stream 3D action recognition approaches simply concatenate or add the geometry and motion features, which cannot well exploit the relationship between two-steam features. To this end, we also propose an improved transformer-based feature fusion module (FFM) to effectively fuse the two-stream features. Based on the proposed FEM and FFM, we build our GMT for 3D action recognition. Extensive experimental results on four benchmark datasets demonstrate the effectiveness of our backbone GMT.

1 January 2024
Verified
Conference Paper
Open access
PDF file
TSTao Sun
Tao Sun
LCLinzheng Chai
Linzheng Chai
JYJian Yang
Jian Yang
YYYuwei Yin
Yuwei Yin
HGHugo Guo
Hugo Guo
ZLZhoujun Li
Zhoujun Li
Tao Sun, Linzheng Chai, Jian Yang, Yuwei Yin, Hongcheng Guo, Jiaheng Liu, Bing Wang, Liqun Yang, Zhoujun Li. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.