Kioxia Develops Image-Recognition System For Logistics
Kioxia Corporation, with partners Tsubakimoto Chain and EAGLYS, has developed an AI-driven image-recognition system for logistics that uses KIOXIA AiSAQ and Memory-Centric AI, and will be demonstrated at the 2025 International Robot Exhibition on December 3–6 in Tokyo. The approach stores extensive product images and features in high-capacity storage and indexes them into SSDs to enable rapid classification of new SKUs without retraining, and AiSAQ is released open-source on GitHub.
Key Points
- 1Introduces AiSAQ image-recognition using memory-centric AI to classify conveyor-belt products without retraining models.
- 2Reduces operational cost and latency by indexing storage-resident features and moving indexes into SSDs.
- 3Enables logistics teams to onboard new SKUs quickly; software released open-source for practitioner adoption.
Scoring Rationale
Official open-source product release and practical demo, strong immediate applicability but represents incremental engineering rather than groundbreaking research.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems