I am looking for exciting computer vision projects to work on - Leave your email and I will be in touch!
Otherwise write me at fml35@cam.ac.uk or connect with me on LinkedIn!
Florian Langer
Computer Vision PhD Student
University of Cambridge
I have recently submitted my PhD thesis, titled "CAD model-based 3D scene reconstruction".
Computer Vision Research Intern
Qualcomm XR Labs
I developed an efficient method for CAD model retrieval and alignment for AR.
Software Engineer - Contracting Work
Apple
I was part of a team developing an exciting new computer vision product.
Research
FastCAD: Real-Time CAD Retrieval and Alignment from Scans and Videos
(ECCV 2024)
Florian Langer, Jihong Ju, Georgi Dikov, Gerhard Reitmayr, Mohsen Ghafoorian
We propose FastCAD, a real-time method that simultaneously retrieves and aligns CAD models for all objects in a given scene in just 50 ms. We achieve high-quality shape retrievals by learning CAD model embeddings in a contrastive learning framework and distilling those into FastCAD.
Sparse Multi-Object Render-and-Compare
(BMVC 2023)
Florian Langer, Ignas Budvytis, Roberto Cipolla
We show how to perform sparse, render-and-compare for multiple objects jointly. This increases both the inference speed as well as the accuracy of the CAD model alignments as natural scene layouts can be learned.
SPARC: Sparse Render-and-Compare for CAD model alignment in a single RGB Image
(BMVC 2022)
Florian Langer, Gwangbin Bae, Ignas Budvytis, Roberto Cipolla
We introduce a sparse, iterative, render-and-compare approach and demonstrate that it is more accurate for CAD model alignment than previous methods. We combine 2D image information including sparse depth and surface normal values with 3D CAD model information and use this as the input to a pose prediction network, SPARC-Net which we train to predict 9 DoF CAD model pose updates.
Leveraging Geometry for Shape Estimation from a Single RGB Image
(BMVC 2021)
Florian Langer, Ignas Budvytis, Roberto Cipolla
We demonstrate how cross-domain key-point matches from an RGB image to a rendered CAD model allows for precise object pose estimation. We further show how the key-point matches can be used to modify the shape of the object. This is important as the accuracy that can be achieved with object retrieval alone is inherently limited to the available CAD models.
Expansion by regions with pySecDec
(Computer Physics Communications 2021)
G. Heinrich, S. Jahn, S.P. Jones, M. Kerner, F. Langer, V. Magerya, A. Poldaru, J. Schlenk, E. Villa
We discuss the technique of expansion by regions from a geometric perspec- tive, and its implementation within pySecDec, a toolbox for the evaluation of dimensionally regulated parameter integrals. The program offers an auto- mated way to perform asymptotic expansions and provides a new mechanism for efficiently evaluating amplitudes, as well as individual integrals. The usage of the new features available within pySecDec is illustrated with several examples.
Learning on the Fly - Precise Positioning of a Drone using Spoken Language Commands
(Master Thesis University of Cambridge 2020)
Florian Langer
I have built a system that allows a user to instruct a drone to fly to arbitrary locations in a room using spoken language commands such as "Fly to the right of the sofa and in front of the table".
Bosonic String Theory and T-Duality
(Master Thesis Imperial College London 2019)
Florian Langer
As string theory may serve as both, a theory of quantum gravity and a unified theory of all fundamental forces, it has naturally been one of the most active research fields of theoretical physics in the last decades. This thesis aims to be a simple introduction to bosonic string theory and one of its beautiful symmetries, T-duality.
Work Experience
Contracting Work Software Engineer - Apple
Cambridge
I was working as part of a team in Cambridge developing an exciting new computer vision product.
April - September 2024
Computer Vision Research Internship - Qualcomm Inc.
Amsterdam
June 2023 - February 2024
As part of my research internship, I developed a system that can accurately estimate a CAD model-based reconstruction of a scene using either a video or depth-sensor input. A CAD-based representation has many advantages, such as being compact, noise-free, and fast to render and simulate, which are crucial for a range of applications in AR. The system I developed was filed with the U.S. Patent and Trademark Office for patenting
Research Internship - Max-Planck Institute for Physics
Munich
I completed a two months summer research project in the phenomenology group. As part of the project I implemented aspects of the expansion by regions methods that allows calculating higher order corrections to loop integrals for particle interactions in Quantum Field Theory.
June - August 2018
Founder of Emote Reality - Virtual Reality Start-Up
London
In Emote Reality we specialised in creating virtual tours for businesses that allow their customers to experience their locations through Virtual Reality glasses or any smart device.
December 2016 - June 2019
Internship at Staramba SE
Berlin
Staramba SE are creating interactive VR experiences. I was working with the level artists in high-poly modelling for constructing an extremely realistic virtual world.
June 2017 - July 2017
Education
PhD in Computer Vision University of Cambridge
Since 2020
I have recently submitted my PhD thesis, titled "CAD model-based 3D scene reconstruction" at the University of Cambridge. I had the pleasure of being supervised by Dr. Ignas Budvytis and advised by Prof. Roberto Cipolla. My thesis introduces various techniques for generating a CAD model-based scene representation from a single image, video or point cloud.
MPhil in Machine Learning and Machine Intelligence - University of Cambridge
I completed the MLMI MPhil with a distinction. My master thesis "Learning on the Fly - Precise Positioning of a Drone using Spoken Language" was supervised by Dr. Ignas Budvytis.
2019 - 2020
MSci & ARCS in Physics with Theoretical Physics
Imperial College London
2015 – 2019
Dean’s list award 2017/2018 and 2018/2019 for top 10% of students in each year cohort.
My Master thesis “Bosonic String Theory and T-duality“ was supervised by Prof. Daniel Waldram.
Abitur
Salem International College
2012 – 2015
I completed the German Abitur with the top grade 1.0 (865 out of 900 points).
Awards: National maths competition “Bundeswettbewerb für Mathematik” 2015 Third Prize, 2014 Official Recognition, “Mathematikolympiade” Federal state final Hessen 2011 7. Place, Hessen 2009 1. Place