Automated vehicle video generation
For the same client behind our vehicle imaging platform, CodeLeap built the system that turns each vehicle's approved photo set into a polished, branded walkaround video automatically — composed in code with Remotion, rendered serverlessly on AWS, and driven by the platform's own computer vision.
What we did CodeLeap built and ran the automated video generation workstream of its multi-year vehicle imaging engagement — a Remotion render API, a slide and effect library, computer-vision-driven highlights, and serverless rendering on AWS Lambda.
Every listing becomes a video. This is the second workstream of the multi-year engagement behind our vehicle imaging platform. Once a vehicle's photos had been captured, cleaned and quality-checked, the client wanted video too — a branded promotional video for every listing, at a volume no editing team could match. CodeLeap built the video system end to end: a rendering engine on Remotion, where every video is composed in React code from a parameterised template; a render API their platform calls for each vehicle; and the AWS infrastructure to produce them at scale. The computer vision itself belongs to the imaging platform — this workstream is what puts it on screen. The client stays unnamed; the capability speaks for itself.
What we built
-
Videos composed in code
Every video is assembled programmatically with Remotion — image, sequence, map and video-in-video slides, a library of label styles and transitions, and zoom behaviour tuned per shot type — so output stays consistent and on-brand, adjusted in code rather than in an editing suite.
-
A render API for the platform
The client's platform and developers request a video through an API — slide types, text, colours, fonts, aspect ratio and a graded style preset — and get back a finished MP4. We wrote the documentation their own team integrated against.
-
Computer-vision-driven effects
The platform's object detection — headlights, wheels, seats — drives animated highlights and zooms that move between a vehicle's actual features, and number plates are blurred or substituted inside the video frames, validated by keypoint detection.
-
Serverless rendering at scale
We benchmarked the engine hard: by 2021 a 24-slide video rendered in around 2.5 minutes on a single EC2 instance — roughly 500 videos a day — before we rebuilt rendering serverlessly on AWS Lambda, with S3 holding assets and DynamoDB tracking every job.
Why this matters to you:
The system went live in production and stayed there, generating listing videos with no editor in the loop, and we supported and evolved it across a multi-year engagement. For the client, video became one more automatic output of the imaging pipeline — the same photos, one more format.
Keep exploring
-
Studio car photos, anywhere. Vehicle imaging platformAn end-to-end vehicle imaging platform for a company in automotive photography. CodeLeap built the capture apps and the computer-vision pipeline that removes backgrounds, builds 360-degree spins and checks every image automatically.
read more -
SBUCodeLeap automated the full sales cycle for a client in the UK energy sector, from lead capture through to signed contract, increasing lead generation capacity by 5x in 3 months.
read more -
WeSnagA large London-based construction company managing major residential sites — dozens of contractors, thousands of defects — came to CodeLeap for a digital snagging platform. Within 6 months, we delivered a fully functioning mobile app that reduced human errors and accelerated the snagging process for construction managers.
read more
