PeopleofSuperintelligence

Researcher

Fisher Yu

Fisher Yu is a computer-vision researcher whose work spans 3D scene representation, dense prediction, autonomous driving and video segmentation. In 2015, he coauthored 3D ShapeNets, the LSUN image dataset and a dilated-convolution context module, followed in 2017 by the SUNCG 3D scene dataset and Deep Layer Aggregation. After completing a PhD at Princeton University and postdoctoral research at UC Berkeley, he joined ETH Zurich in December 2020 and coauthored BDD100K, SAM-PT, UniDepth and DVPSFormer. As of October 2026, he is an assistant professor at ETH Zurich and directs the Visual Intelligence and Systems group in its Computer Vision Lab.

Now at
Computer Vision Lab
Works
168
Citations
40,953
h-index
58

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2015

    3D ShapeNets and the LSUN dataset

    In 2015, he coauthors 3D ShapeNets, a deep volumetric representation that supports joint object recognition and shape completion from 2.5D depth maps. In June 2015, he and his coauthors construct LSUN, a dataset with around one million labeled images across 10 scene categories and 20 object categories.

  2. November 2015

    Dilated convolutions for dense prediction

    With Vladlen Koltun in November 2015, he presents a dilated-convolution module that aggregates multi-scale contextual information for dense prediction without losing resolution.

  3. 2017

    SUNCG and Deep Layer Aggregation

    His 2017 semantic scene completion paper introduces SUNCG, a dataset of synthetic 3D scenes with dense volumetric annotations for single-depth-image training. In July 2017, he and his coauthors present Deep Layer Aggregation, which iteratively and hierarchically merges the feature hierarchy for higher accuracy with fewer parameters.

  4. 2019–2020

    Joining ETH Zurich

    While he is a postdoctoral researcher at UC Berkeley, ETH Zurich appoints him in July 2019 as a tenure-track assistant professor of computer vision. He holds a PhD from Princeton University and joins ETH Zurich's Department of Information Technology and Electrical Engineering in December 2020.

  5. 2020

    BDD100K for driving research

    For autonomous-driving evaluation in 2020, he and his coauthors construct BDD100K, a dataset with 100,000 driving videos and 10 tasks for studying image-recognition algorithms.

  6. July 2023

    Segment Anything meets point tracking

    He coauthors SAM-PT in July 2023, a point-centric interactive video-segmentation method that combines Segment Anything with long-term point tracking.

  7. 2024

    UniDepth and monocular depth

    The CVPR 2024 Highlight paper UniDepth, which he coauthors in 2024, reconstructs metric 3D scenes from single images across domains.

  8. As of October 2026

    Assistant professor at ETH Zurich

    In July 2026, he coauthors DVPSFormer, a unified online architecture for efficient 4D scene understanding in autonomous driving. As of October 2026, he is an assistant professor at ETH Zurich and directs the Visual Intelligence and Systems group in its Computer Vision Lab.

Career

  1. Since 2019ETH ZurichTenure-track Assistant Professor of Computer Vision
  2. NowComputer Vision LabDirector
  3. 2017–2019University of California, BerkeleyPostdoctoral Researcher

Education

  1. Princeton UniversityPhD

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes ETH Zurich. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes ETH Zurich. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  3. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes ETH Zurich. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  4. Universities had a net outflow of 2,068 moves in AI careersIncludes ETH Zurich. Our estimates of career flows and median h-index across universities and companies in AI, including the role of joint appointments and earlier academic jobs.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

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

  1. 3D ShapeNets: A Deep Representation for Volumetric Shapes openaccess.thecvf.com. Checked 22 October 2026.
  2. LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop arxiv.org. Checked 22 October 2026.
  3. Multi-Scale Context Aggregation by Dilated Convolutions arxiv.org. Checked 22 October 2026.
  4. Semantic Scene Completion From a Single Depth Image openaccess.thecvf.com. Checked 22 October 2026.
  5. Deep Layer Aggregation arxiv.org. Checked 22 October 2026.
  6. Thirteen new professors appointed ethz.ch. Checked 22 October 2026.
  7. New CLS faculty members at ETH Zurich learning-systems.org. Checked 22 October 2026.
  8. Fisher Yu | NCCR Automation nccr-automation.ch. Checked 22 October 2026.
  9. Welcome to D-ITET, Prof. Fisher Yu! ee.ethz.ch. Checked 22 October 2026.
  10. BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning openaccess.thecvf.com. Checked 22 October 2026.
  11. Segment Anything Meets Point Tracking arxiv.org. Checked 22 October 2026.
  12. UniDepth: Universal Monocular Metric Depth Estimation | Luigi Piccinelli lpiccinelli-eth.github.io. Checked 22 October 2026.
  13. DVPSFormer: Efficient Online Depth-aware Video Panoptic Segmentation for Autonomous Driving arxiv.org. Checked 22 October 2026.