This lab works on phones, but writing code is much easier on a laptop.
Perception · Classical Feature Detection and Matching · P4-L1 · Intermediate
A robot wants to track how the world moves between two camera frames. It needs points it can recognise again: not flat walls, not plain edges, but corners.
You will tune a corner detector, write the corner score, and select a clean set of points that appear in both images.
You have passed every stage. Now connect it back to the theory and move on:
Plain Python with math (and slamkit for small matrix helpers: matmul, transpose, solve, inverse, identity). Images, point clouds and detections arrive as lists inside the dictionary D; an image is a list of rows, each row a list of pixels. There is no NumPy or OpenCV: you write the algorithm yourself, which is the point of the lab. Print with print().
Stage 1 is a tuning experiment judged by repeatability across two views. Stage 2 writes the Harris response from image gradients. Stage 3 selects corners by threshold and non-maximum suppression. Your code is graded on hidden image pairs; the simulator keeps the true corners and the true transform private.
Educational simplification: simulated 64 by 48 grey images of overlapping rectangles seen from two viewpoints related by a small rotation, a shift, a brightness change and Gaussian noise of 6 grey levels. Real images have texture, blur, scale change and clutter. All images in this lab are simulated.
Get notified when new robotics labs are released. One email per new lab, nothing else.