Computer Vision with Neural Networks
Find the matching picture
BOTIE
Examples, experiments and workspace controls
Connect a tool's inputs to see what it produces. You can try a branch before connecting it to Result.
Try your own inputs
Use a copy of the example to try your own data. Keep the problem's assumptions in mind, and predict the effect of a change before checking the output. You can also add another branch to compare two ways of doing the calculation.
Context for this challenge
Find the closest stored image vector, then return its image ID. Keep the descriptors separate from the identifiers they point to.
Botie's query can be compared with stored image vectors, but the user needs the identity of the retrieved picture rather than its position in the vector table.
Measure distances to stored descriptors, find the closest position, and retrieve the aligned image ID.
Examples and results
- Query image vector
- 00.9
- 11.1
- Stored image vectors
0 0 1 1 2 0 - Stored image IDs
- 012
- 134
- 256
- Required result
34
Try a fresh construction
This clears this draft, including its undo history. Earlier successful attempts stay recorded. Build again without opening hints to record an independent attempt.
Stuck? Inspect a worked construction
The tools below form one possible solution. You can assemble it, follow its signals, and undo that change to return to your own construction.
Assumptions and test precision
This is a small construction problem. The lesson explains the assumptions and the parts of the full method that this example leaves out. Numerical tests allow a difference of 0.001. Lists and structures must also have the right shape.
Two stored pictures are equally close
This check uses the pieces from “Carry the nearest vector's image ID into the result”.
BOTIE
- Query image vector
- 00
- 10
- Stored image vectors
1 0 -1 0 0 3 - Stored image IDs
- 091
- 117
- 244
Use a number, text, or an array matching the requested result. For a matrix, put each row in brackets. Saving your prediction runs your current construction on these inputs and checks the earlier examples too.
Build a construction in the workspace above first. This case will test those pieces.
Why could a mathematically nearest descriptor still retrieve the wrong subject for a user?
Retrieval quality depends on what the representation makes similar, as well as on the distance search.
This is a small construction problem. The lesson explains the assumptions and the parts of the full method that this example leaves out.
Work through the Python challenges →