Updates: 2020/02/21: Add pixel-to-graphIn autonomous driving, convolutional neural networks are the go-to tool for various perception tasks. Although CNNs are great at distilling information from camera images (or a sequence of them in form of a video clip), I constantly bump into all kinds of metadata that do not lend themselves to convolutional neural networks.
![Convolutional Neural Networks with Heterogeneous Metadata](https://cdn-ak-scissors.b.st-hatena.com/image/square/ba96d18fb917ceb6fc440f6950ffacff66dc5ae2/height=288;version=1;width=512/https%3A%2F%2Fmiro.medium.com%2Fv2%2Fresize%3Afit%3A500%2F1%2AgY04VqBWECGjfz5mRlGemA.png)