Note: All images in this directory, unless specified otherwise, are licensed under CC BY-NC 4.0.
| Figure number | Description |
|---|---|
| 8-1 | Sample of Clarifai’s results |
| 8-2 | Sample of Microsoft Cognitive Services results |
| 8-3 | Sample of Google Cloud Vision’s results |
| 8-4 | Sample of Amazon Rekognition’s results |
| 8-5 | Sample of IBM Watson’s Visual Recognition results |
| 8-6 | Sample of Algorithmia’s style transfer results |
| 8-7 | A cost comparison of different cloud-based vision APIs |
| 8-8 | WER for different text extraction APIs as of August 2019 |
| 8-9 | Averaged faces among different gender and skin tone, from Pilot Parliaments Benchmark (PPB) |
| 8-10 | Face detection comparison across APIs, tested in April and May 2017 on the PPB |
| 8-11 | Face detection comparison across APIs in August 2018 on the PPB, conducted by Inioluwa Deborah Raji et al. |
| 8-12 | Image-tagging API performance on geographically diverse images from the Dollar Street dataset |
| 8-13 | Average accuracy (and standard deviation) of six cloud APIs versus income of the household where the images were collected |
| 8-14 | Creating a new project in Custom Vision |
| 8-15 | Uploading images on CustomVision.ai |
| 8-16 | The Train button in the upper-right corner of the CustomVision.ai page |
| 8-17 | Relative precision and recall for our sample training set with 200 images per class |
| 8-18 | Some of the possible tags returned by the API |
| 8-19 | A chart showing the F1 score for custom classifier services, as of August 2019 (higher is better) |
| 8-20 | Effect of varying size of training data per class on test F1 score (higher is better) |
| 8-21 | Effect of resizing an image on agreement rate and file size reduction relative to the original image |
| 8-22 | Effect of compressing an image on agreement rate and file size reduction relative to the original image |
| 8-23 | The Uber Drivers app prompts the driver to take selfie to verify the identity of the driver |
| 8-24 | Giphy extracts text from animations as metadata for searching |
| 8-25 | Detections made by InDro Robotics |
| 8-26 | Drone identifies two stranded swimmers and releases an inflatable pod that they cling onto |
