Xiaofeng Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiaofeng Huang has a total of 102 co-authors. Here are the top co-authors by number of shared publications:
- Haibing Yin
Hangzhou Dianzi University
14 shared publications Haibing Yin
7 shared publicationsYang Zhou
School of Communication Engineering, Hangzhou Dianzi University, Hangzhou, China
5 shared publications- Yang Zhou
Hangzhou Dianzi University, Hangzhou, China
5 shared publications - Changcai Lai
School of Electronics and Information Engineering, Hangzhou Dianzi University, Hangzhou, China
3 shared publications Chenggang Yan
Hangzhou Dianzi Univ, Sch Commun Engn, Hangzhou 310015, Peoples R China
3 shared publications
Latest Publications
- 8 November 2026
- Published by Springer Nature
- p. 3-16
GRACE: Boosting Video MLLMs with Grounded Action-Centric Evidence for Viewer Sentiment Prediction
- 8 November 2026
- Published by Springer Nature
- p. 3-16
GRACE: Boosting Video MLLMs with Grounded Action-Centric Evidence for Viewer Sentiment Prediction
- Open Access
- 16 September 2026
- Published by Elsevier in Journal of Visual Communication and Image Representation
- Vol. 121
360VFI: 360-degree video frame interpolation
- 16 September 2026
- Published by Elsevier in Journal of Visual Communication and Image Representation
- Vol. 121
Open Access360VFI: 360-degree video frame interpolation
- 15 September 2026
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 23 (5) , 184
Lightweight MLP via structural-aware U-Net knowledge distillation for fast CU partition in AVS3 intra coding
- 15 September 2026
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 23 (5) , 184
Lightweight MLP via structural-aware U-Net knowledge distillation for fast CU partition in AVS3 intra coding
- Open Access
- 12 June 2026
- Published by Elsevier in SSRN Electronic Journal
GhostIC: A Ghost-Driven Lightweight Network for Efficient Learned Image Compression
Learned image compression (LIC) models have achieved remarkable performance but typically rely on computationally expensive backbones, which limits their applicability on resource-constrained embedded devices. To address this limitation, we present GhostIC, a Ghost-driven lightweight image compression network that achieves efficient inference while maintaining competitive rate–distortion performance. In GhostIC, we design a GhostBlock inspired by GhostNet where redundant features are generated from intrinsic features using inexpensive operations. This design integrates prior knowledge of image spatial redundancy, enabling efficient representation for image compression. Furthermore, the GhostBlock integrates the decoupled fully connected (DFC) attention mechanism, which enhances long-range dependency modeling with minimal overhead and compensates for the representational limitations of lightweight convolutions. Extensive experiments on standard benchmarks demonstrate that GhostIC delivers a rate-distortion performance on par with VVC intra coding while substantially reducing computational complexity compared to state-of-the-art learned models. With its favorable balance between efficiency and performance, GhostIC provides a practical solution for real-time learned image compression on resource-constrained devices. The source code will be made available upon publication.
- 12 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessGhostIC: A Ghost-Driven Lightweight Network for Efficient Learned Image Compression
Learned image compression (LIC) models have achieved remarkable performance but typically rely on computationally expensive backbones, which limits their applicability on resource-constrained embedded devices. To address this limitation, we present GhostIC, a Ghost-driven lightweight image compression network that achieves efficient inference while maintaining competitive rate–distortion performance. In GhostIC, we design a GhostBlock inspired by GhostNet where redundant features are generated from intrinsic features using inexpensive operations. This design integrates prior knowledge of image spatial redundancy, enabling efficient representation for image compression. Furthermore, the GhostBlock integrates the decoupled fully connected (DFC) attention mechanism, which enhances long-range dependency modeling with minimal overhead and compensates for the representational limitations of lightweight convolutions. Extensive experiments on standard benchmarks demonstrate that GhostIC delivers a rate-distortion performance on par with VVC intra coding while substantially reducing computational complexity compared to state-of-the-art learned models. With its favorable balance between efficiency and performance, GhostIC provides a practical solution for real-time learned image compression on resource-constrained devices. The source code will be made available upon publication.
- 3 April 2026
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 20 (4) , 239
Autoencoder-based 3D Model Pretraining for No-Reference Point Cloud Quality Assessment
- 3 April 2026
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 20 (4) , 239
Autoencoder-based 3D Model Pretraining for No-Reference Point Cloud Quality Assessment
- 2 December 2025
- Published by Elsevier in Neurocomputing
- Vol. 666
Enhancing graph contrastive learning with knowledge graph embedding for recommendation
- 2 December 2025
- Published by Elsevier in Neurocomputing
- Vol. 666
Enhancing graph contrastive learning with knowledge graph embedding for recommendation
- 29 November 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 170
Robust ESLOM-based direction finding for noncircular sources via sparse nested arrays in α-stable noise environments
- 29 November 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 170
Robust ESLOM-based direction finding for noncircular sources via sparse nested arrays in α-stable noise environments
- 2 November 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (6) , 1-14
Hierarchical all-zero block early determination algorithm for low-complexity versatile video coding
- 2 November 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (6) , 1-14
Hierarchical all-zero block early determination algorithm for low-complexity versatile video coding
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10642-10651
Sparse4DGS: Flow-Geometry Assisted 4D Gaussian Splatting for Dynamic Sparse View Synthesis
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10642-10651
Open AccessSparse4DGS: Flow-Geometry Assisted 4D Gaussian Splatting for Dynamic Sparse View Synthesis