Compute geolocation from object shadows!
summary
Claim: compute a rough geolocation from object shadow lengths using the algorithm from the paper Determining the Geographical Location of Image Scenes Based on Object Shadow Lengths by Frode Eika Sandnes. Setup uses cardboard, a pen, and a 2‑megapixel camera and requires careful vertical mounting and constant camera height. Pipeline extracts shadow length from three images at different times, maps each to a latitude/longitude circle, and intersects them to get candidates visualized with MATLAB plots. Errors come from tilt, weather, and manual shadow measurement.
AI-written summary · about
chapters (7)
- 0:00Intro
- 1:51ELEVATION OF THE SUN
- 2:26EQUIPMENT
- 3:15CHALLENGES
- 5:16LATITUDE AND LONGITUDES CALCULATION
- 6:56PLOT USING MATLAB
- 8:47OBSERVATIONS AND FUTURE IMPROVEMENTS
related talks
12mEclipseTouch: Touch Segmentation on Ad Hoc Surfaces using Worn Infrared Shadow Casting
01h 00mCloud and Shadow Recognition in Satellite Imagery
38mShadows Generation in the Wild
10mSo THAT’S how my phone knows where I am!
13mBuild location-based AR experiences at world scale
37mWhere the Heck Am I? Understanding and Managing Location in Your Application
10mMapping sidewalks and street furniture with AI and computer vision
06mThe Crowd Wisdom for Location Privacy of Crowdsensing Photos: Spear or Shield?