PeopleofSuperintelligence

Researcher

Chen-Yu Wang

Also known as Chen Wang

Chen-Yu Wang is a computer-vision researcher whose work spans face tracking, sign-language synthesis, clinical gait analysis, driving vision-language models and avatar animation. He coauthored a facial-landmark filtering method in 2023, the SynthSL image-synthesis paper in 2024, and a ChatGPT-4o gait-assessment study in 2025. In 2026, he coauthored a driving-model benchmark, a clinical gait dataset and graph-attention work for real-time avatar animation. As of October 2026, he is a researcher in DFKI's Augmented Vision department.

Now at
DFKI

Timeline

  1. 2023

    3D face landmark tracking

    In 2023, he coauthors “Confidence-Aware Clustered Landmark Filtering for Hybrid 3D Face Tracking” with Jilliam Maria Diaz Barros, Didier Stricker and Jason Raphael Rambach. Its frame-to-frame module uses grouped-landmark Kalman filters and improves landmark accuracy under large occlusion, extreme head poses and blurriness.

  2. May 2024

    SynthSL for sign language

    In May 2024, he coauthors “SynthSL: Expressive Humans for Sign Language Image Synthesis” with Jilliam Maria Diaz Barros, Jameel Malik, Abdalla Elsayed Abdou Elsayed Mohamed Arafa and Didier Stricker. The Augmented Vision group gets two papers accepted for FG 2024, including SynthSL.

  3. 2025

    GPT-4o gait assessments

    In 2025, he coauthors a study that uses ChatGPT-4o to generate full-body gait assessments from spatiotemporal motion data. A blinded evaluation finds its assessments comparable to human-generated assessments in correctness and clarity.

  4. 2026

    FutureVQA for driving

    In 2026, he coauthors a study that tests driving vision-language models for response consistency and temporal reasoning. The paper introduces FutureVQA, a human-annotated benchmark for future-scene reasoning.

  5. August 2026

    A clinical gait dataset

    In August 2026, he coauthors a data descriptor that combines motion-capture records with expert clinical gait assessments. The dataset includes recordings from 23 patients; 15 have both preoperative and six-week postoperative sessions.

  6. August 2026

    Graph attention for avatars

    In 2026, he coauthors IK-GAT, a graph-attention network that recovers full-body joint rotations from tracked 3D positions in one pass. Its animation-ready local rotations can drive a rigged avatar or provide pose parameters for SMPL-like body models.

  7. As of October 2026

    Researcher at DFKI

    As of October 2026, he is a researcher in DFKI's Augmented Vision department.

  1. Nonprofit roles are 2.3 times as common in AI safetyIncludes DFKI. We estimate where people focused on safety, alignment and evaluation work. Nonprofit roles appear for 25.1% of safety people with known affiliations, compared with 11.0% in other areas.
  2. The top 10 organizations employed 22% of people shaping AI in 2025: 51% of company researchers and 21% of university researchersIncludes DFKI. We estimate the share of people shaping AI who worked at the top 10 organizations each year from 1955 to 2025, comparing concentration across companies, universities, and institutes and labs.
  3. The United States employs an estimated 48% of people shaping AIIncludes DFKI. We estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes DFKI. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.

Sources and credits

Photo: We have no published photo of Chen-Yu Wang yet. Send us a free-licensed photo.

  1. CONFIDENCE-AWARE CLUSTERED LANDMARK FILTERING FOR HYBRID 3D FACE TRACKING av.dfki.de. Checked 22 October 2026.
  2. FG2024: 2 papers accepted av.dfki.de. Checked 22 October 2026.
  3. Exploring Large Language Models for Automated Gait Analysis av.dfki.de. Checked 22 October 2026.
  4. Probing the Reliability of Driving VLMs: From Inconsistent Responses to Grounded Temporal Reasoning openaccess.thecvf.com. Checked 22 October 2026.
  5. A Multimodal Dataset for Clinical Gait Analysis compromising Optical Motion Capture and Textual Gait Analyses from Patients undergoing Total Knee Arthroplasty nature.com. Checked 22 October 2026.
  6. IK-GAT | Real-Time Inverse Kinematics with Graph Attention · Saif Khan saifkhichi.com. Checked 22 October 2026.
  7. Chen-Yu Wang av.dfki.de. Checked 22 October 2026.