Unverified50% confidenceFactExact time
JST-2's characters can come from an image or a piece of text, and the motions are built in.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/29/2026
First Seen
Valid until: 1/27/2027
Sources
Related Entities
Related Claims
UnverifiedThe author believes that image models are renderers, not the source of truth for motion; motion should first be obtained from a real source (motion capture or live-action video) and then rendered with any character.59% similarUnverifiedReference-driven generation extracts information such as motion, composition, and character poses from the original footage, then re-renders it in a new visual style.56% similarUnverifiedSuperpose offers pose suggestions rather than directly generating composite photos, thereby preserving the user's agency in taking real photographs.56% similarUnverifiedA scene graph is a structured representation of image content in the form of a directed graph, with objects as nodes and inter-object relationships as edges.55% similarUnverifiedKling and Pika support describing camera motion through text prompts55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1111327API
curl https://kongchang.com/api/v1/knowledge/claims/1111327MCP
get_claim(id=1111327)