Life after DLS for a Computer Vision researcher

AI Discussions
,
1

I’ve been working as a “junior” computer vision researcher for almost a year now.
Most of my learning was On Job Training on my own. The DLS really helped me make sense of what I’m doing and eventually to improve my performance.

I’m now looking for a course or book that will take me to the next level in CNNs. This means, as an example:

  1. Vision transformers
  2. Different loss functions for CNN models
  3. How inference works when the crop size is different from the actual images we’re getting?
  4. Deeper dive into TensorFlow and PyTorch
  5. Latest research in the field and what people do to improve their models

Any other thing that I haven’t thought of.
Your feedback would be warmly welcomed

2

Hi fabiansc,

Here’s my two cents.

After you have completed relevant courses at deeplearning.ai, the next step is to dive into the literature yourself. There are many advanced ideas you can find in books, articles, and courses elsewhere (e.g. huggingface, the tensorflow or pytorch sites, ArXiv, O’Reilly, Packt, and so on and so forth). It may also help to follow people on social media, to go to informal meetings, to participate in online sessions, to ask questions on stackoverflow, and so on, and so forth. Nobody has all the answers, and everyone in AI is trying to figure out how to keep up with developments that are going so fast. The good part: you’ll never get bored!