Problem Description and Pipeline
- Text detection
- Character segmentation
- Character classification
Go out and collect large training sets of positive and negative examples.
Take that green rectangle and we slide it over a bit and then run that new image patch through our classifier to decide if there’s a pedestrian there.
Getting Lots of Data and Artificial Data
- artificial data synthesis
- just collect the data and you label it yourself
- crowd sourcing
Ceiling Analysis What Part of the Pipeline to Work on Next
Where should you allocate resources?
Which of these boxes is most worth your efforts, trying to improve the performance of.
- choose one module
- provide it the correct text detection outputs
- And then, use the same evaluation metric as before, to measure what is the overall accuracy of the entire system
- go to step 1, but choose another one