Collaborative Edge-Cloud Face Recognition

The problem

Improve face-recognition inference performance by using edge and cloud computing collaboratively.

My contribution

• Co-author “Optimizing face recognition inference with a collaborative edge–cloud network.”
• Contribute to research on distributed face-recognition inference.

Technical approach

The research investigates collaborative edge–cloud inference rather than treating device and cloud processing as isolated alternatives. The linked paper is the source for the experimental method and reported results.

1. Research question

   Investigate collaborative face-recognition inference across edge and cloud computing.

2. Engineering considerations

   Account for the relationship between device computation and network communication when interpreting inference performance.

3. Reproducible evidence

   Consult the linked Sensors paper for the experimental setup and results. The published work is a team contribution.

Engineering considerations

Inference placement introduces a balance between device computation and network communication. Hardware and network conditions matter when interpreting throughput or latency results.

Outcome

Peer-reviewed publication in Sensors, volume 22, issue 21, article 8371 (2022).

Project resources