Multimodal Vision Research Laboratory

MVRL

Research Area: Geometric Vision

How do we recover geometry, depth, and 3D structure from images and panoramas? Geometric vision supports scene understanding, generative modeling, and geospatial applications across the lab. Recent work includes open-world stereo generation with unsupervised matching (GenStereo), consistent text-to-360 scene generation (PanoDreamer), generative-free 3D scene recovery for occlusion removal (DeclutterNeRF), and omnidirectional depth via stereo matching from multi-cylindrical panoramas (MCPDepth). We also study vanishing-point detection, top-view reasoning for outdoor scenes, and calibration cues from natural phenomena such as horizons and rainbows.

View All Publications →

All Publications

  1. Thumbnail for MCPDepth: Omnidirectional Depth Estimation via Stereo Matching from Multi-Cylindrical Panoramas
    Qiao F, Xiong Z, Zhu X, Ma Y, He Q, Jacobs N. 2026. MCPDepth: Omnidirectional Depth Estimation via Stereo Matching from Multi-Cylindrical Panoramas. In: Omnidirectional Computer Vision Workshop (OmniCV) (IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshop).
    paper bibtex Transportation Geometry
  2. Luo Y, Qiao F, Xiong Z, Li Y, Jacobs N. 2026. GenOpticalFlow: A Generative Approach to Unsupervised Optical Flow Learning.
    bibtex GenAI Geometry
  3. Thumbnail for Towards Open-World Generation of Stereo Images and Unsupervised Matching
    Qiao F, Xiong Z, Xing E, Jacobs N. 2025. Towards Open-World Generation of Stereo Images and Unsupervised Matching. In: IEEE/CVF International Conference on Computer Vision (ICCV).
    code website paper bibtex GenAI Geometry
  4. Xiong Z, Chen Z, Li Z, Xu Y, Jacobs N. 2025. PanoDreamer: Consistent Text to 360 Scene Generation. In: 4th Computer Vision for Metaverse Workshop (IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops).
    paper bibtex GenAI Geometry
  5. Liu W, Xiong Z, Li X, Jacobs N. 2025. DeclutterNeRF: Generative-Free 3D Scene Recovery for Occlusion Removal. In: 4th Computer Vision for Metaverse Workshop (IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops).
    paper bibtex GenAI Geometry
  6. Thumbnail for Detecting Vanishing Points using Global Image Context in a Non-Manhattan World
    Zhai M, Workman S, Jacobs N. 2016. Detecting Vanishing Points using Global Image Context in a Non-Manhattan World. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). DOI: 10.1109/CVPR.2016.610.
    code paper doi bibtex Calibration Geometry
  7. Workman S, Greenwell C, Zhai M, Baltenberger R, Jacobs N. 2015. DeepFocal: A Method for Direct Focal Length Estimation. In: IEEE International Conference on Image Processing (ICIP). DOI: 10.1109/ICIP.2015.7351024.
    website paper doi bibtex Calibration Geometry
  8. Thumbnail for Scene Shape Estimation from Multiple Partly Cloudy Days
    Workman S, Souvenir R, Jacobs N. 2015. Scene Shape Estimation from Multiple Partly Cloudy Days. Computer Vision and Image Understanding (CVIU):116–129. DOI: 10.1016/j.cviu.2014.10.002.
    website doi bibtex Calibration Geometry
  9. Thumbnail for A Pot of Gold: Rainbows as a Calibration Cue
    Workman S, Mihail RP, Jacobs N. 2014. A Pot of Gold: Rainbows as a Calibration Cue. In: European Conference on Computer Vision (ECCV). 820–835. DOI: 10.1007/978-3-319-10602-1_53.
    website paper doi bibtex Calibration Geometry
  10. Abrams A, Hawley C, Miskell K, Stoica A, Jacobs N, Pless R. 2013. Shadow Estimation Method for "The Episolar Constraint: Monocular Shape from Shadow Correspondence". arXiv preprint 1304.4112 [cs.CV].
    paper bibtex Geometry Calibration