Share Your Requirements
Nantabor
LANCERProject Description
Dan Meyer from Nantabor approached Oodles to transform static real estate property images into short 3–5 second AI-driven video clips. The objective was to enhance property listings by integrating realistic human activity, motion, and environmental context across scenes like dining, poolside, and outdoor settings to improve visual engagement and user experience.
Scope Of Work
The project focused on converting static images into animated clips by integrating AI-generated characters and environmental elements. This included scene interpretation, character placement, motion design, and realism alignment. It also involved handling multiple scenarios such as dining setups, pool interactions, and fireworks, along with iterative refinements based on feedback on motion quality, object interaction, and visual consistency.
Our Solution
Our solution for this project focused on executing within the constraints of AI-generated motion and realism limitations:
- Interpreted loosely defined inputs and translated static property visuals into dynamic scenes with contextual human presence
- Integrated AI-generated characters by aligning lighting, scale, shadows, and perspective with existing environments
- Built multi-layered animations across dining, poolside, and outdoor scenes to simulate realistic usage scenarios
- Identified and corrected inconsistencies such as unnatural object overlap, motion imbalance, and environmental mismatches
- Refined character movement speed, behaviour, and scene pacing through iterative feedback cycles to improve realism
- Tested multiple AI models and motion variations to stabilize outputs where natural interactions were difficult to achieve
- Managed execution despite incorrect initial assets and access issues without disrupting delivery flow
- Communicated technical constraints clearly when further refinements reached limits of current AI capabilities