endorsedOngoing

Unity Shopper Simulation for Autonomous Checkout

Services Covered:

Custom Software Development • Game Development

Industry

Software Development

Duration

55 months

Budget

50K - 199K

Start Date

1 January 2022

Team Size

1-5 Employees

Engagement Model

Offshore

Project Capability Score (PCS)

77/100

Good

Project Capability Score (PCS) estimates how capable N-iX Games is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

Standard AI

San Francisco, United States of America

Project Summary

N-iX Game & VR Studio built a Unity-based simulation tool for an autonomous checkout company to render videos of simulated shoppers for ML training and evaluation. The vendor created systems to generate randomized shopper trajectories, randomize store layouts and product placement, and produce segmentation and depth maps for rendered outputs. The team (2 developers and 1 project manager) has been making steady progress, meeting milestone estimates and tracking work in Jira.

Key Challenges

Build a simulation tool to render videos of simulated shoppers moving and shopping naturally
Generate a wide variety of random situations and scenarios for robustness
Produce data suitable for training and evaluating machine learning models

Project Deliverables

Unity-based shopper simulation platform
Agent trajectory generator for simulated shoppers
Randomized store layout, product arrangement, and texture systems
Segmentation maps and depth maps produced during rendering

Project Solution

The vendor implemented the project using the Unity game engine. They built a system to generate random trajectories for artificial agents so shoppers move around a virtual store and take/place items according to generated scripts. They implemented functionality to randomize store layouts, product arrangements, and textures, and they produced detailed segmentation and depth maps during rendering. The engagement included 2 developers and 1 project manager.

Project Outcome

The team has made steady progress toward deliverables
The team has been accurate with time estimates for intermediate milestones
The vendor is organized and tracks progress in Jira with periodic demos
The client is satisfied and confident the team can continue delivering features

Platforms

AR/VR/XR

Tech Stack

Unity
Jira

Client Endorsement

Overall Review Rating

4.38

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“Thanks to N-iX Game & VR Studio's efforts, the client has seen steady progress toward deliverables. The team has been accurate with their timelines when it comes to milestones. They are organized to track progress through Jira and accommodate the internal team's questions and concerns.”
Anonymous
Machine Lrarning Lead
Date: 05/09/2022