Xi Peng, Ph.D., computer science

Welcome! I am a tenure-track faculty at the University of Delaware (Since 2019):

I am leading Deep-REAL (Deep Robust and Explainable AI Lab).

Research interest: Machine Learning, Deep Learning, and Computer Vision. [Google Scholar]

Email: xipeng at udel dot edu Tel: (302) 831-3185

Office: 440 Smith Hall, 18 Amstel Ave, Newark, DE 19716


[NSDI'21] Adapting Wireless Mesh Network Configuration from Simulation to Reality via Deep Learning-based Domain Adaptation. [PDF]

[CVPR'21] Uncertainty-guided Model Generalization to Unseen Domains. [PDF] [60s-Video] [Code]

[CVPR'21 Oral] Learning View-Disentangled Human Pose Representation by Contrastive Cross-View Mutual Information Maximization. [PDF]

[AAAI'21] Multimodal learning with severely missing modality. [PDF] [60s-Video] [15m-Video] [Code]

[ICLR'21 Spotlight] A good image generator is what you need for high-resolution video synthesis. [PDF]

[NeurIPS'20] Maximum-entropy adversarial data augmentation for improved generalization and robustness. [PDF]

[CVPR'20] Learning to learn single domain generalization. [PDF] [60s-Video] [Code]

[CVPR'20] Knowledge as priors: Cross-modal knowledge generalization for datasets without superior knowledge. [PDF] [60s-Video]

[NeurIPS'19] Semantic-guided multi-attention localization for zero-shot learning. [PDF]

[NeurIPS'19] Rethinking kernel methods for node representation learning on graphs. [PDF]

[ICCV'19 Oral] AdaTransform: Adaptive Data Transformation. [PDF]

[CVPR'19] Semantic graph convolutional networks for 3d human pose regression. [PDF]


  • May 2021, [Grant] Our interdisciplinary team (PI Prof. Tian-Jian Hsu) was awarded an intellectual neighborhood accelerator grant. Thanks to COE.

  • Mar 2021, [Media] Check the UDaily story of our $1M interdisciplinary research supported by NSF!!

  • Mar 2021, [NSDI'21] AI + IoT "Adapting Wireless Mesh Network Configuration from Simulation to Reality". [PDF]

  • Mar 2021, [Service] Dr. Peng served on NSF Review Panel.

  • Mar 2021, [Grant] Dr. Peng (PI) was awarded a research grant for AI-based biomechanical analysis. Thanks to CDC/NIOSH!!

  • Mar 2021, [CVPR'21] "Uncertain Out-of-Domain Generalization" (Robust/Explainable AI). Congrats to Fengchun (2-year PhD)! [PDF]

  • Mar 2021, [CVPR'21 (Oral) rate 4.3%] "View-disentangled Human Pose Representation" (Human-centered Vision). Congrats to Long! [PDF]

  • Feb 2021, [ICLR'21] "High-resolution Video Synthesis" (Efficient AI). Congrats to Yu! [PDF]

  • Feb 2021, [Grant] Dr. Peng (PI) was awarded a search grant for self-supervised multimodal learning. Thanks to Snap Research!

  • Jan 2021, [Grant] our interdisciplinary team (PI Prof. Julie Klinger) secured a 5-year (2021-2026) research grant on AI/ML-based geoscience research. Thanks to NSF!!

  • Dec 2020, a new interdisciplinary PhD position on AI/ML/DL-based Geospatial Data Analytics is available [Link].

  • Dec 2020, [AAAI'21] "Multimodal Learning with Severely Missing Modality" (Efficient/Explainable AI) is accepted by. Congrats to Mengmeng (1-year PhD)!! [PDF/Code]

  • Sep 2020, [NeurIPS'20] "Maximum-Entropy Adversarial Data Augmentation" (Robust AI) is accepted by. Congrats to Long!

  • Sep 2020, [Service] Dr. Peng served on NSF Review Panel.

  • May 2020, [Service] Dr. Peng served as the Guest Editor of the CVIU special issue on "Action Recognition and Detection". [Call for submissions]

  • Mar 2020, [IJCV'20] "Semantic Guided Motion Forecasting and Video Generation" (Human-centered Vision).

  • Feb 2020, [Travel] Dr. Peng attended CRA Career Workshop, Washington, D.C.

  • Feb 2020, [CVPR'20] "Single Domain Generalization" (Robust/Explainable AI). Congrats to Fengchun (1-year PhD)!! [PDF/Code]

  • Feb 2020, [CVPR'20] "Cross-Modal Knowledge Generalization" (Robust AI). Congrats to Long! [PDF]

  • Dec 2019, [Travel] we present two papers on NeurIPS'19, Vancouver, Canada.

  • Dec 2019, [Grant] Dr. Peng (PI) received a research grant for multimodal learning. Thanks to Snap Research!

  • Oct 2019, [Travel] present our work as an Oral paper on ICCV'19, Seoul, Korea.

  • Sep 2019, [Grant] Dr. Peng (PI) was awarded a seed grant for transfer learning. Thanks to Data Science Institute!

  • Sep 2019, [NeurIPS'19] "Kernel Methods for Node Representation Learning on Graphs" (Explainable AI) is accepted by. [PDF/Code]

  • Sep 2019, [NeurIPS'19] "Semantic-Guided Multi-Attention Localization for Zero-shot Learning" (Efficient AI) is accepted by. [PDF/Code]

  • Aug 2019, [Service] Dr. Peng will serve as invited as (Senior) Program Committee of AAAI '20, NeurIPS'20, ICLR'20, CVPR'20.

  • Aug 2019, [Travel] attend KDD'19, Anchorage, Alaska.

  • July 2019, [ICCV'19 (Oral) rate 4.7%) "Adaptive Data Augmentation" (Robust AI). [PDF/Code]

  • July 2019, [BMVC'19] "Construct Dynamic Graphs (GCN) for Hand Gesture Recognition" (Human-centered Vision). [PDF/Code]

  • Jun 2019, [Travel] Dr. Peng attend CVPR'19, Los Angeles, California.

  • May 2019, [KDD'19 (Oral) rate 9.2%)] "Scalable Global Alignment Graph Kernel". [PDF]

  • Apr 2019, [Service] Dr. Peng served as Session Chair of ICIG'19 special session on "Learning toward Visual Recognition in the Wild". [Call for Papers]

  • Mar 2019, [TPAMI (Impact factor: 17.73)] "Quantized U-Net" (Efficient AI). [PDF/Code]

  • Mar 2019, [Travel] Dr. Peng visited Orthopedics Department at New Jersey Medical School, Newark, New Jersey.

  • Feb 2019, [CVPR'19] "Semantic Graph Convolutional Network (GCN) for 3D Human Pose Estimation" (Human-centered Vision). [PDF/Code]

  • Jan 2019, [Journal of Biomechanics] "Deep Learning-based Biomechanics Analysis" (Human-centered Vision). [PDF]

  • Jan 2019, [IEEE THMS] "Predicting Back Joint Load in Lifting" (Human-centered Vision). [PDF]

  • Jan 2019, [Service] Dr. Peng served as (Senior) Program Committee of NIPS'19 & ICLR'19 & CVPR'19 & ICCV'19 & AAAI '19 & IJCAI'19 & ACMMM'19.