Featured MEdia
Supply Chains

Evercoast Acquires Depthkit to Advance AI Spatial Video

The combined company creates the most comprehensive suite of technologies to accelerate generative AI for 3D spatial video.

Evercoast, a leading provider of end-to-end 3D spatial video software, has announced the acquisition of Depthkit, a pioneer in volumetric capture. This strategic acquisition marks a significant advancement in the quest to capture and generate 3D digital humans and spatial video, which enables viewers to experience subjects and dynamic scenes from all angles across various digital platforms. Furthermore, this milestone represents a crucial step in the evolution of generative AI, transitioning from 2D to 3D creation, as the extensive datasets amassed by both companies will address a critical shortfall in the necessary training data.

Historically, comprehensive 3D datasets have been scarce due to the complexity and cost associated with their creation, resulting in algorithms lagging behind their 2D counterparts. AI platforms developed by OpenAI, Meta, Google, Microsoft, Anthropic, and others have relied on expansive sources of text and 2D datasets, leaving a gap in 3D spatial data. The merger of Evercoast and Depthkit seeks to fill this void by uniting the largest customer base for 3D volumetric video globally. Together, these companies are positioned to deliver a highly scalable software platform capable of capturing and generating thousands of assets within a single day. This method is significantly faster and costs 20 times less than multi-camera arrays employing traditional processing pipelines and expensive hardware.

“Teaming up with Evercoast signifies an exciting new beginning for Depthkit and the entire 3D content creation community,” remarked James George, CEO of Scatter, the company behind Depthkit. “By collaborating, we are not only merging two cutting-edge technologies but also aligning our visions to create tools that are innovative and accessible. We are confident that our combined efforts will stretch the boundaries of spatial video and unlock new opportunities worldwide.”

Spatial video datasets and models address unique characteristics that 2D video fails to replicate effectively at scale. Real-world variability in human motion, influenced by factors such as lighting, occlusions, and diverse environments, necessitates datasets with multiple camera perspectives for comprehensive representations. Multi-camera setups further enhance the capture of depth information, crucial for accurately understanding spatial relationships between body parts and clothing. Moreover, training models using data from various camera views facilitates better generalization than monocular or synthetic datasets, particularly in complex interactions involving other objects or individuals.

The increasing demand for specialized AI development in fields such as human performance capture, medical training, industrial simulation, robotics, sports performance, and higher education signals a vast, addressable market for this technology. VFX and other creative technologists are also utilizing volumetric data in conjunction with machine learning and AI video, revealing significant overlaps between the two companies’ verticals. Their combined technology stack not only enhances the market but provides a hardware-agnostic solution.

Moving forward, existing Depthkit customers will receive full access to their current features in addition to numerous enhanced functionalities available through Evercoast. For further information regarding Evercoast’s offerings for spatial video production, please visit evercoast.com.

Persons: James George
Company Names: Evercoast, Depthkit, Scatter, OpenAI, Meta, Google, Microsoft, Anthropic
Titles: Depthkit, Evercoast

Disclaimer: This article has been auto-generated from a syndicated RSS feed and has not been edited by Vitrina staff. It is provided solely for informational purposes on a non-commercial basis.