I recently gave a talk at the 2019 Deep Learning Summit in Boston about the nuances of building modern visual search and recommendation systems at Salesforce. There were many strong talks from many companies and independent research labs:

The rework summit comprised three core themes:

• Infrastructures and Frameworks

• Adversarial Learning for Text-to-Image Synthesis

• Recommender Systems and Forecasting Demand

I spoke about one of our many projects at Einstein - a Visual Search API, which consists of an ensemble of deep metric learned models for product recognition. Our convnets are trained to learn compact manifolds in latent spaces for retrieving consumer products. I added a link to the video of the talk above and the slides are embedded below, note that the slides with animated content were not recorded/rendered. So I added them below!

Abstract: Fine-grain recognition remains an unsolved problem and in the general case it may even be as difficult as self-driving cars. Many technical challenges remain in achieving accurate production-level image retrieval at web scale (handling catalogs of tens of millions of items). This talk details the steps and highlights the hurdles in building such a search platform. At Commerce Cloud Einstein, we have developed a custom multi-stage pipeline of deep metric learning models for product detection and recognition. Our networks are trained to discover a manifold representing the space of all consumer products. We will present the current architectures in our embedding networks, i.e. the mapping from consumer images to the product feature space, as well as the most promising research directions. Implementation level details will be covered insofar as they make efficient fine-grain retrieval possible, and performance evaluation (both statistical as well as qualitative) measures will be described.

videos and animations: