Cheng Deng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Cheng Deng has a total of 266 co-authors. Here are the top co-authors by number of shared publications:
- Dacheng Tao
Nanyang technological university, Singapore, Singapore
25 shared publications - Xu Yang
Xidian University, Xi'an, China
21 shared publications - Heng Huang
Department of Computer Science, University of Maryland, College Park, College Park, MD, USA
16 shared publications - Muli Yang
Institute for Infocomm Research (I2R), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore
15 shared publications - Kun Wei
School of Electronic Engineering, Xidian University, Xian, China
14 shared publications Xu Yang
School of Electronic Engineering Xidian University China
11 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 12404-12413
A Turn Toward Better Alignment: Few-Shot Generative Adaptation with Equivariant Feature Rotation
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 12404-12413
Open AccessA Turn Toward Better Alignment: Few-Shot Generative Adaptation with Equivariant Feature Rotation
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15128-15137
RCTR: Risk-Calibrated Token Routing for Long-Horizon Online Video Inpainting
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15128-15137
Open AccessRCTR: Risk-Calibrated Token Routing for Long-Horizon Online Video Inpainting
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4973-4981
Hierarchical Temporal Residual Dynamics Network for Spiking Object Detectors
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4973-4981
Open AccessHierarchical Temporal Residual Dynamics Network for Spiking Object Detectors
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 7400-7409
COME: Advancing Motion Representation and Generative Modeling for High-Quality Text-to-Motion
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 7400-7409
Open AccessCOME: Advancing Motion Representation and Generative Modeling for High-Quality Text-to-Motion
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (11) , 15252-15266
Non-Negative Tensor Factorization Based Bi-Clustering on Anchor Graphs
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (11) , 15252-15266
Non-Negative Tensor Factorization Based Bi-Clustering on Anchor Graphs
- Open Access
- 3 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
Streaming Video Moment Retrieval via Temporal Forecast Diffusion Model
With the advancement of online media, the demand for real-time interaction with streaming video has rapidly increased, highlighting the importance of Streaming Video Moment Retrieval (SVMR). This task aims to evaluate the relevance between streaming video and textual queries in real-time. Unlike offline videos, streaming videos must be processed on-the-fly, which presents substantial challenges due to the absence of future frames and the redundancy of historical frames. These factors lead to insufficient contextual information, which hinders reasoning about the current frame. To address these challenges, we propose DiffuStream, a novel approach from a generative perspective. Specifically, DiffuStream leverages a Temporal Forecast Diffusion Model to efficiently synthesize future semantics conditioned on observable contexts, thereby providing additional future information to assist in reasoning about the current frame. Complementarily, we introduce a Segment-based Dual Compressor to extract multi-granularity cues from noisy history while serving as a robust condition to assist future generation. By integrating the synthesized future with the refined history, DiffuStream constructs a comprehensive historical-present-future context, facilitating precise cross-modal reasoning. Extensive experiments on the ActivityNet Captions, TACoS, and MAD datasets validate the superior effectiveness of DiffuStream.
- 3 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
Open AccessStreaming Video Moment Retrieval via Temporal Forecast Diffusion Model
With the advancement of online media, the demand for real-time interaction with streaming video has rapidly increased, highlighting the importance of Streaming Video Moment Retrieval (SVMR). This task aims to evaluate the relevance between streaming video and textual queries in real-time. Unlike offline videos, streaming videos must be processed on-the-fly, which presents substantial challenges due to the absence of future frames and the redundancy of historical frames. These factors lead to insufficient contextual information, which hinders reasoning about the current frame. To address these challenges, we propose DiffuStream, a novel approach from a generative perspective. Specifically, DiffuStream leverages a Temporal Forecast Diffusion Model to efficiently synthesize future semantics conditioned on observable contexts, thereby providing additional future information to assist in reasoning about the current frame. Complementarily, we introduce a Segment-based Dual Compressor to extract multi-granularity cues from noisy history while serving as a robust condition to assist future generation. By integrating the synthesized future with the refined history, DiffuStream constructs a comprehensive historical-present-future context, facilitating precise cross-modal reasoning. Extensive experiments on the ActivityNet Captions, TACoS, and MAD datasets validate the superior effectiveness of DiffuStream.
- 10 September 2026
- Published by Springer Nature
- p. 55-72
Multi-Anchor Distillation with Text-Guided Analytic Classifier for Continual Learning
- 10 September 2026
- Published by Springer Nature
- p. 55-72
Multi-Anchor Distillation with Text-Guided Analytic Classifier for Continual Learning
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (16) , 24135-24143
High Circular Dichroism Chiral Metasurface for Mid-Infrared Molecular Fingerprint Sensing via Hybrid Fano-Phonon Resonances
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (16) , 24135-24143
High Circular Dichroism Chiral Metasurface for Mid-Infrared Molecular Fingerprint Sensing via Hybrid Fano-Phonon Resonances
- 3 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. PP (01628828) , 1-16
Deciphering Object Concepts: Hierarchical Cross-Modal Relational Reasoning for Mining Object-Attribute-Affordance Associations
Object Concept Learning (OCL) aims to recognize high-level attributes and affordances of objects and to infer the causal relationships between them. The key is to accurately model the many-to-many mapping between objects and concepts: While an object may possess multiple concepts, a concept can also belong to multiple objects. Existing methods primarily rely on attention mechanisms to capture label correlations, which limits their ability to comprehend high-level concepts and to perform effective causal reasoning. Inspired by the human cognitive process of progressive understanding, a Hierarchical Cross-Modal Relational Reasoning (CORE) framework is proposed to enhance the understanding of object concepts through hierarchical interaction and reasoning between visual and textual modalities. Specifically, a coarse-to-fine relational reasoning module is developed, in which multi-step learnable prompts are employed to progressively localize the conceptual information of objects, thereby improving the accuracy of object-concept mapping. Subsequently, to facilitate the modeling of causal relationships between object attributes and affordances, a counterfactual reasoning mechanism is introduced. By constructing counterfactual samples and distinguishing the predictive outputs of factual and counterfactual parts, the model's ability to capture causality among concepts is enhanced. Significant performance gains and extensive visualization analysis demonstrate the superiority of our method.
- 3 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. PP (01628828) , 1-16
Deciphering Object Concepts: Hierarchical Cross-Modal Relational Reasoning for Mining Object-Attribute-Affordance Associations
Object Concept Learning (OCL) aims to recognize high-level attributes and affordances of objects and to infer the causal relationships between them. The key is to accurately model the many-to-many mapping between objects and concepts: While an object may possess multiple concepts, a concept can also belong to multiple objects. Existing methods primarily rely on attention mechanisms to capture label correlations, which limits their ability to comprehend high-level concepts and to perform effective causal reasoning. Inspired by the human cognitive process of progressive understanding, a Hierarchical Cross-Modal Relational Reasoning (CORE) framework is proposed to enhance the understanding of object concepts through hierarchical interaction and reasoning between visual and textual modalities. Specifically, a coarse-to-fine relational reasoning module is developed, in which multi-step learnable prompts are employed to progressively localize the conceptual information of objects, thereby improving the accuracy of object-concept mapping. Subsequently, to facilitate the modeling of causal relationships between object attributes and affordances, a counterfactual reasoning mechanism is introduced. By constructing counterfactual samples and distinguishing the predictive outputs of factual and counterfactual parts, the model's ability to capture causality among concepts is enhanced. Significant performance gains and extensive visualization analysis demonstrate the superiority of our method.
- 22 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35 (10577149) , 4253-4268
Counterfactual Risk Minimization for Out-of-Distribution Generalization
The out-of-distribution (OOD) property in data is deemed as one main challenge hindering the generalization ability of machine learning algorithms. However, the underlying reasons for this property remain an intriguing and open question that has yet to be fully understood. In this paper, we seek to enhance our understanding of the OOD phenomenon by framing it as a problem of distribution shift and addressing it through two complementary causal perspectives. The first is a generative causal view that elucidates the data generation process. We introduce a novel three-dimensional coordinate system to represent three fundamental distribution shifts, illustrating their role in various OOD generalization problems. The second is an anti-causal view that focuses on the model learning process. We develop an effective approach dubbed Counterfactual Risk Minimization (CRM) to address arbitrary distribution shifts in a unified framework. Additionally, we introduce a new multidomain visual recognition dataset called CONA to facilitate further exploration of OOD generalization. We conduct evaluations of CRM alongside several state-of-the-art competitors on four benchmark datasets under the three distribution shifts. The results not only affirm CRM’s superiority but also shed light on potential future directions.
- 22 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35 (10577149) , 4253-4268
Counterfactual Risk Minimization for Out-of-Distribution Generalization
The out-of-distribution (OOD) property in data is deemed as one main challenge hindering the generalization ability of machine learning algorithms. However, the underlying reasons for this property remain an intriguing and open question that has yet to be fully understood. In this paper, we seek to enhance our understanding of the OOD phenomenon by framing it as a problem of distribution shift and addressing it through two complementary causal perspectives. The first is a generative causal view that elucidates the data generation process. We introduce a novel three-dimensional coordinate system to represent three fundamental distribution shifts, illustrating their role in various OOD generalization problems. The second is an anti-causal view that focuses on the model learning process. We develop an effective approach dubbed Counterfactual Risk Minimization (CRM) to address arbitrary distribution shifts in a unified framework. Additionally, we introduce a new multidomain visual recognition dataset called CONA to facilitate further exploration of OOD generalization. We conduct evaluations of CRM alongside several state-of-the-art competitors on four benchmark datasets under the three distribution shifts. The results not only affirm CRM’s superiority but also shed light on potential future directions.