Abstract
We introduce a novel Deep Network architecture that implements the full feature point handling pipeline, that is, detection, orientation estimation, and feature description. While previous works have successfully tackled each one of these problems individually, we show how to learn to do all three in a unified manner while preserving end-to-end differentiability. We then demonstrate that our Deep pipeline outperforms state-of-the-art methods on a number of benchmark datasets, without the need of retraining.
References
Field Guide to Northern Tree-related Microhabitats: Descriptions and size limits for their inventory in boreal and hemiboreal forests of Europe and North America
Swiss Federal Institute for Forest, Snow and Landscape Research WSL, Switzerland, 2024.Data Champions Lunch Talks – Green Bytes: Data-Driven Approaches to EPFL Sustainability
Data Champions Lunch Talks – Green Bytes: Data-Driven Approaches to EPFL Sustainability, EPFL, CM 1 221, April 18, 2024.Comparison of Three Viral Nucleic Acid Preamplification Pipelines for Sewage Viral Metagenomics
Food and Environmental Virology. 2024. DOI : 10.1007/s12560-024-09594-3.How to Support Students to Develop Skills that Promote Sustainability
Teaching Transversal Skills for Engineering Studens: A Practical Handbook of Activities with Tangibles; EPFL, 2024.How to Support Students Giving Each Other Constructive Feedback, Especially When It Is Difficult to Hear
Teaching Transversal Skills for Engineering Studens: A Practical Handbook of Activities with Tangibles; EPFL, 2024.How teachers can use the 3T PLAY trident framework to design an activity that develops transversal skills
Teaching Transversal Skills for Engineering Studens: A Practical Handbook of Activities with Tangibles; EPFL, 2024.The conceptual foundations of innate immunity: Taking stock 30 years later
Immunity. 2024-04-09. Vol. 57, num. 4, p. 613-631. DOI : 10.1016/j.immuni.2024.03.007.Radio-Activities: Architecture and Broadcasting in Cold War Berlin
Cambridge, MA; London: MIT Press, 2024.No Last One
Revue Matières. 2024. num. 18.All That is Solid
Transcalar Prospects in Climate Crisis; Zurich: Lars Müller, 2024.Teaser Video
This teaser video shows feature matching results with our integrated LIFT pipeline and SIFT, for selected sequences of all three datasets, Strecha, DTU, and Webcam. Our results are significantly better overall compared to SIFT. Note that, in our experiments, SIFT still gives results that are on par with the state-of-the-art when evaluated as a whole pipeline. Please see the paper for details.
Supplementary material
Click the following link for the supplementary appendix for implementation details.
Dataset and Codes
Datasets used in the paper.
Codes for LIFT with learned modules.
Contacts
Kwang Moo Yi | [e-mail] |
Eduard Trulls | [e-mail] |
Vincent Lepetit | [e-mail] |
Pascal Fua | [e-mail] |