Edge Impulse is a platform focused on deploying ML models to low-power edge devices and embedded systems. It supports both vision and other models like audio, time-series, and signal processing. Edge Impulse is uniquely good at working withmicrocontrollers and has SDKs for single-board computers and mobile devices.
The design focus on TinyML makes it less suited for high-resource, general-purpose tasks like video processing and running modern, state-of-the-art ML models. It also requires some familiarity with embedded systems. It typically requires custom coding your application logic to run on the embedded board.
Chose Edge Impulse if: you're working on an IoT or wearable device that's not capable of running more powerful models, framework, and logic.
If you're deeply engrained in the Tensorflow ecosystem and want to deploy avariety of Tensorflow models in different modalities like NLP, recommender systems, and audio in addition to CV models, Tensorflow Serving may be a good choice.
It can be complex to setup and maintain and lacks features many users would consider table stakes (like pre- and post-processing which in many cases will need to be custom coded). Like several of the other servers listed here, it lacks depth in vision-specific functionality.
Choose Tensorflow Serving if: the Tensorflow ecosystem is very important to you and you're willing to put in the legwork to take advantage of its advanced feature set.
Inference turns any computer or edge device into a command center for your computer vision projects.