Image Matching: Local Features and Beyond CVPR 2019 Workshop: June - - PowerPoint PPT Presentation

▶
image matching local features and beyond
SMART_READER_LITE
LIVE PREVIEW

Image Matching: Local Features and Beyond CVPR 2019 Workshop: June - - PowerPoint PPT Presentation

Apr 30, 2023 329 likes •502 views

Image Matching: Local Features and Beyond CVPR 2019 Workshop: June 16 (morning) Vassileios Balntas (Scape), Vincent Lepetit (U. Bordeaux), Johannes Schnberger (Microsoft), Eduard Trulls (Google), Kwang Moo Yi (U. Victoria) Organizers

slide-1
SLIDE 1

Image Matching: Local Features and Beyond

CVPR 2019 Workshop: June 16 (morning)

Vassileios Balntas (Scape), Vincent Lepetit (U. Bordeaux), Johannes Schönberger (Microsoft), Eduard Trulls (Google), Kwang Moo Yi (U. Victoria)

slide-2
SLIDE 2

Organizers

Vassileios Balntas

Scape Technologies

Vincent Lepetit

  • U. Bordeaux

Johannes Schönberger

Microsoft

Eduard Trulls

Google

Kwang Moo Yi

  • U. Victoria
slide-3
SLIDE 3

8:45 - 9:00 Welcome 9:00 - 9:30 Amir Zamir (Stanford/UC Berkeley) Collection of Large-scale Densely-labeled 3D Data from the Real World Without a Single Click 9:30 - 10:15 Jiri Matas (CTU Prague) On the Art of Establishing Correspondence 10:15 - 11:00 Coffee Break + Poster Session 11:15 - 12:00 Torsten Sattler (Chalmers U. of Technology, Gothenburg) In Defense of Local Features for Visual Localization 12:00 - 12:15 IMW2019 Challenge 12:15 - 12:30 Zixin Luo (HKUST) Winner of the Phototourism Challenge 12:30 - 12:45 Challenge results and awards

Program

https://image-matching-workshop.github.io/

slide-4
SLIDE 4

Focal point: image matching

  • Matching rigid scenes across baselines, time, weather, etc.
  • Underlying technologies common to key CV/ML problems: mapping,

re-localization, SLAM, augmented & virtual reality, autonomous navigation, robotics, etc.

slide-5
SLIDE 5

Google Maps AR

slide-6
SLIDE 6

Scape Localisation Engine

SOSNet Oral+Poster

slide-7
SLIDE 7

A classical problem, but far from solved… "IRL "

Environmental changes Large baselines Occlusions Local vs. World-Scale

slide-8
SLIDE 8

Structured methods remain king

Understanding the Limitations of CNN-based Absolute Camera Pose Regression. Sattler et al., CVPR 2019.

More at 11:15!

slide-9
SLIDE 9

The last bastion?

slide-10
SLIDE 10

The last bastion?

Us

slide-11
SLIDE 11

The last bastion?

Us Deep Learning Deep Learning Deep Learning More Deep Learning

slide-12
SLIDE 12

To be clear...

  • We use machine learning a lot

○ But not end-to-end

  • We don't know if individual components generalize well

○ Does performance translate down-stream? ○ Are we focusing on the right problems?

slide-13
SLIDE 13

Topics (not limited to)

Novel modalities (e.g. pano vs aerial)

(Cross-view geo-localization, CVPR'19)

Learning feature extractors

D2-Net, CVPR'19

Adversarial methods

(CycleGAN, ICCV'17)

Learning feature matchers

Learning Correspondences, CVPR'18

slide-14
SLIDE 14

Phototourism Challenge

Eduard Trulls (Google) Kwang Moo Yi (U. Victoria) Sri Raghu Malireddi (U. Victoria) Yuhe Jin (U. Victoria)

slide-15
SLIDE 15

SILDa Challenge

Vassileios Balntas (Scape)

slide-16
SLIDE 16

8:45 - 9:00 Welcome 9:00 - 9:30 Amir Zamir (Stanford/UC Berkeley) Collection of Large-scale Densely-labeled 3D Data from the Real World Without a Single Click 9:30 - 10:15 Jiri Matas (CTU Prague) On the Art of Establishing Correspondence 10:15 - 11:00 Coffee Break + Poster Session 11:15 - 12:00 Torsten Sattler (Chalmers U. of Technology, Gothenburg) In Defense of Local Features for Visual Localization 12:00 - 12:15 IMW2019 Challenge 12:15 - 12:30 Zixin Luo (HKUST) Winner of the Phototourism Challenge 12:30 - 12:45 Challenge results and awards

Program

https://image-matching-workshop.github.io/