Warehouse Localizer Model (Beta)

AI Data Capture SDK 3.2

Overview

Model Name: warehouse-localizer

As part of Zebra’s Frontline AI Enablers, this beta release introduces the new Warehouse Localizer Model, expanding our suite of enterprise-ready AI solutions with powerful new capabilities. The Warehouse Localizer model is designed to detect key assets within warehouses and retail environments. It identifies items such as boxes, pallets, warehouse shelves, and shelf labels. This model is ideal for use cases like automated inventory counting and asset tracking.

Integrated with the Zebra AI Data Capture SDK Localizer building block, the model allows developers to incorporate warehouse asset detection directly into their applications. This beta release provides stable core functionality for evaluation and integration testing, with ongoing refinements planned prior to general availability.

Note on Beta Features: Features marked as (Beta) are under active development and may be subject to change. They are made available for evaluation and feedback purposes. We do not recommend using Beta features in production environments.

Version History

New in v1.0.4

  • Improved detection accuracy for boxitem and pallet classes.

Requirements

  • Operating System: Android 14 or higher. For specific OS versions, refer to AI Data Capture SDK Release Notes from the Zebra support portal.

  • Minimum SDK Version: AI Data Capture SDK 3.0.2 or later.

  • Supported Zebra Devices:

    Features Platform Device Model
    Products with DSP
    Fastest and most battery efficient    		 QC6490 TC53, TC58, TC73, TC78, ET60, ET65

    For more information on devices based on platform, see Zebra Platform Devices.

Technical Details

  • Input Resolution Recommended: 832x832

Note: For input resolutions beyond this recommendation, the model load time could be higher.​

Performance Guidance

Model Default Input
Image Resolution		 SNPE Load Time (ms) Inference Time (ms) API Time (SDK + Inference) (ms) Inference Memory Usage (MB)
warehouse-localizer 832x832 26 19 981 63

Measurements Notes:

  • All measurements are estimations​.
  • Measurements were conducted using the Zebra TC53 device equipped with the Qualcomm 6490 chipset operating on the DSP AI accelerator​.
  • Images were 4MP Resolution under 300 Lux​ lighting conditions.
  • Measurements were based on a UPCA 13 mil barcode.
  • Estimates reflect detection times only and do not include decode time.

Resources