A system and method for receiving an ordered set of images and analyzing the images to determine at least one position in space and at least one motion vector in space and time for at least one object represented in the images is disclosed. Using these vectors, a four dimensional model of at least a
A system and method for receiving an ordered set of images and analyzing the images to determine at least one position in space and at least one motion vector in space and time for at least one object represented in the images is disclosed. Using these vectors, a four dimensional model of at least a portion of the information represented in the images is formulated. This model generally obeys the laws of physics, though aberrations may be imposed. The model is then exercised with an input parameter, which, for example, may represent a different perspective than the original set of images. The four dimensional model is then used to produce a modified set of ordered images in dependence on the input parameter and optionally the set of images, e.g., if only a portion of the data represented in the images is modeled. The set of images may then be rendered on a display device.
대표청구항▼
1. A method, comprising: receiving at least one stream of input images representing a physical scene having three spatial dimensions and a temporal dimension, representing a movement of at least two natural objects in the physical scene, at least one of the at least two natural objects being alive;p
1. A method, comprising: receiving at least one stream of input images representing a physical scene having three spatial dimensions and a temporal dimension, representing a movement of at least two natural objects in the physical scene, at least one of the at least two natural objects being alive;processing at least the at least one stream of input images, with at least one automated processor, to generate a spatial and temporal computer model of the at least two natural objects, moving and interacting according to a set of intrinsic parameters describing at least a first motion of the at least two natural objects in the physical scene, comprising both first independent motion and first contact between the at least two natural objects, consistent with the laws of physics comprising at least conservation of mass, conservation of momentum, and conservation of energy of the at least two natural objects;receiving a set of extrinsic parameters which are distinct from the set of intrinsic motion parameters, the set of extrinsic parameters describing at least a second motion of the at least two natural objects interacting in the physical scene, comprising both second independent motion and second contact between the at least two natural objects, the first independent motion being different from the second independent motion and the first contact being different from the second contact;receiving at least one statistic for a random variable associated with an outcome of the second motion and the second contact, comprising a likelihood of an injury to the at least one of the at least two natural objects which is alive, as a result of at least the set of extrinsic parameters;rendering the spatial and temporal computer model, with the at least one automated processor, to define a spatial and temporal representation of the at least two natural objects within the physical scene according to at least the set of extrinsic parameters, portions of the set of intrinsic parameters consistent with the set of extrinsic parameters, and a performance statistic for a predefined activity represented in the defined spatial and temporal representation; andgenerating a rendered output of the animated computer model, with the at least one automated processor, as at least one stream of output images defining at least two spatial dimensions and a temporal dimension. 2. The method according to claim 1, wherein the computer model is generated based on both the at least one stream of input images and an external database of characteristics of the at least one of the at least two natural objects. 3. The method according to claim 1, further comprising receiving the set of extrinsic parameters, by acquiring images through an input device comprising a camera. 4. The method according to claim 1, wherein said processing at least the at least one stream of input images comprises: creating a respective three dimensional spatial model of each of the at least two natural objects;extracting information representing non-hidden surfaces of the at least two natural objects in the at least one stream of input images;extrapolating information representing hidden surfaces of the at least two natural objects in the at least one stream of input images; andtexture mapping surfaces of the respective three dimensional spatial models of the at least two natural objects based on the extracted information representing non-hidden surfaces in the at least one stream of input images and the extrapolated information representing the hidden surfaces. 5. The method according to claim 1, further comprising receiving a set of extrinsic parameters which are distinct from the set of intrinsic motion parameters, as movement information received through an input device comprising at least one of a an accelerometer and a gyroscope. 6. The method according to claim 1, wherein: the physical scene comprises a sports arena in which a competitive sports event occurs, having an interaction outcome;the at least two natural objects comprise at least one first human athlete competing against at least one second human athlete within the sports arena;the at least one stream of input images comprises televised video images of the at least one first human athlete competing against at least one second human athlete within the sports arena; andfurther comprising receiving a human user input comprising the set of extrinsic parameters to alter the interaction outcome. 7. The method according to claim 1, wherein: the at least natural objects comprise at least two human athletes competing according to a game having a set of rules;further comprising analyzing the at least one stream of input images to determine a compliance with the set of rules. 8. The method according to claim 1, wherein the generated rendered output comprises a stereoscopic video presentation. 9. The method according to claim 1, wherein the at least two natural objects in the physical scene comprise: at least one human athlete competing in a sports event; andat least one sports projectile object which interacts with the at least one human athlete as a part of the sports event. 10. A system for processing a stream of images, comprising: a first input port configured to receive at least one stream of input images representing a physical scene having three spatial dimensions and a temporal dimension, representing a movement of at least two natural objects in the physical scene;at least one automated processor, configured to: process at least the at least one stream of input images to generate a spatial and temporal computer model of the at least two natural objects, moving and interacting according to a set of intrinsic parameters describing at least a first motion of the at least two natural objects in the physical scene, comprising both first independent motion and first contact between the at least two natural objects, consistent with the laws of physics comprising at least conservation of mass, conservation of momentum, and conservation of energy of the at least two natural objects;extract the set of intrinsic parameters;receive a set of extrinsic parameters from a second input port, which are distinct from the set of intrinsic motion parameters, the set of extrinsic parameters describing at least a second motion of the at least two natural objects interacting in the physical scene, comprising both second independent motion and second contact between the at least two natural objects, the first independent motion being different from the second independent motion and the first contact being different from the second contact;determine at least one statistic for a random variable associated with an outcome of the second independent motion and the second contact, comprising a likelihood of damage to at least one of the at least two natural objects, as a result of movement according to at least the set of extrinsic parameters;render the spatial and temporal computer model, to define a spatial and temporal representation of the at least two natural objects within the physical scene according to at least the set of extrinsic parameters, portions of the set of intrinsic parameters consistent with the set of extrinsic parameters, and a performance statistic for a predefined activity represented in the defined spatial and temporal representation; andgenerate a rendered output of the animated computer model as at least one stream of output images defining at least two spatial dimensions and a temporal dimension; andan output port configured to communicate the generated rendered output. 11. The system according to claim 10, wherein the at least one automated processor is further configured to generate the computer model based on both the at least one stream of input images and information received through the Internet from an external database of characteristics of the at least two natural objects. 12. The system according to claim 10, wherein: the physical scene comprises a sports arena in which a competitive sports event occurs, having an interaction outcome;the at least two natural objects comprise at least two human athletes competing within the sports arena;the at least one stream of input images comprises televised video images of the at least two human athletes competing within the sports arena;the second input port is configured to receive a human user input comprising the set of extrinsic parameters to cause the at least one automated processor to alter the interaction outcome of the at least two human athletes competing within the sports arena represented in the generated rendered output. 13. The system according to claim 10, wherein: the at least two natural objects comprise a sports projectile and at least two human athletes competing according to a game having a set of rules involving the sports projectile;wherein the at least one automated processor is further configured to analyze the at least one stream of input images to determine a compliance with the set of rules of the game. 14. A method, comprising: receiving a first stream of input images representing a three dimensional physical scene over time, representing a movement and contact of at least two natural objects in the physical scene, at least one of the at least two natural objects comprising a sports competitor of a competitive game represented in the first stream of input images;generating, with at least one automated processor, a spatial and temporal model of each of the at least two natural objects, the model reflecting laws of physics comprising at least conservation of mass, conservation of momentum, and conservation of energy;extracting, with the at least one automated processor, a set of intrinsic parameters describing the movement and contact of the at least two natural objects in the physical scene, consistent with the generated spatial and temporal model of each of the at least two natural objects;receiving a set of extrinsic parameters which differ from the set of intrinsic motion parameters;applying, with the at least one automated processor, the set of extrinsic parameters and the at least one statistic for the random variable, to the generated spatial and temporal model of each of the at least two natural objects, to represent a modification of the movement and contact of the at least two natural objects in the physical scene;rendering, with the at least one automated processor, the generated spatial and temporal model of each of the at least two natural objects, to define a modified representation of the three dimensional scene over time according to at least the modification of the movement and contact of the at least two natural objects in the physical scene;outputting a second stream of output images comprising the rendered generated spatial and temporal model of each of the at least two natural objects defining the modified representation of the three dimensional scene over time; andreceiving at least one statistic for a random variable associated with an outcome of the modification of the movement and contact of the at least two natural objects in the physical scene, comprising a likelihood of injuries to the sports competitor represented in the second stream as a result of at least movement of the at least two natural objects in the physical scene according to the set of extrinsic parameters. 15. The method according to claim 14, wherein the at least one statistic for the random variable associated with the outcome of the movement and contact of the at least two natural objects in the physical scene comprises a likelihood of completion of a sports objective of a competitive game represented in the second stream. 16. The method according to claim 14, wherein the competitive game represented in the second stream is football. 17. The method according to claim 14, further comprising receiving an interactive advertisement from the Internet for presentation in the second stream. 18. The method according to claim 14, further comprising receiving the set of extrinsic parameters by acquiring images of a human user through an input device comprising a camera. 19. The method according to claim 14, further comprising receiving the set of extrinsic parameters by sensing movements of a human user through an input device comprising at least one of an accelerometer and a gyroscope.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (53)
Steven M. Hoffberg ; Linda I. Hoffberg-Borghesani, Adaptive pattern recognition based control system and method.
Cosatto, Eric; Graf, Hans Peter; Potamianos, Gerasimos; Schroeter, Juergen, Audio-visual selection process for the synthesis of photo-realistic talking-head animations.
Barbour, Blair A.; Stilwell, Richard J.; Stilwell, Jason M., Automated collection, processing and use of sports movement information via information extraction from electromagnetic energy based upon multi-characteristic spatial phase processing.
Nash,Shawn P., Computer animation of simulated characters using combinations of motion-capture data and external force modelling or other physics models.
Hoffberg Steven Mark ; Hoffberg-Borghesani Linda Irene, Ergonomic man-machine interface incorporating adaptive pattern recognition based control system.
Minear, Kathleen; Pooley, Donald; Smith, Anthony O'Neil, Fusion of a 2D electro-optical image and 3D point cloud data for scene interpretation and registration performance assessment.
Moezzi Saied ; Katkere Arun ; Jain Ramesh, Immersive video, including video hypermosaicing to generate from multiple video views of a scene a three-dimensional v.
Ingrid B. Carlbom ; Yves D. Jean ; Sarma V G K Pingali, Method and apparatus for deriving novel sports statistics from real time tracking of sporting events.
Neely, III, Howard; Azuma, Ronald T.; Isdale, Jerry; Daily, Mike, Method and apparatus for real-time group interactive augmented-reality area monitoring, suitable for enhancing the enjoyment of entertainment events.
Jiang,Dalong; Zhang,Hong Jiang; Zhang,Lei; Yan,Shuicheng; Hu,Yuxiao, Method and system for constructing a 3D representation of a face from a 2D representation.
Chu,Stephen Mingyu; Goel,Vaibhava; Marcheret,Etienne; Potamianos,Gerasimos, Method for likelihood computation in multi-stream HMM based speech recognition.
Jain Ramesh C. ; Hicks Terry Randolph ; Bailey Asquith A. ; McKinley Ryan B. ; Kuramura Don Yamato ; Katkere Arun L., Multi-perspective viewer for content-based interactivity.
Barton James M. ; McInnis Roderick James ; Moskowitz Alan S. ; Goodman Andrew Martin ; Chow Ching Tong ; Kao Jean Swey, Multimedia time warping system.
de Cuetos, Philippe; Iyengar, Giridharan R.; Neti, Chalapathy V.; Potamianos, Gerasimos, System and method for microphone activation using visual speech cues.
Abdallah,Mohammad; Chennupaty,Srinivas; Dreyer,Robert S.; Julier,Michael A.; Kong,Katherine; Mennemeier,Larry; Thakkar,Ticky S., System and method for performing an insert-extract instruction.
Cavallaro Richard H. ; Gepner Jerry N. ; Gloudemans James R. ; Honey Stanley K. ; Squadron William F. ; White Marvin S., System for determining the position of an object.
Gee, Will; Thompson, David; Taylor, Murray; Deriso, Martin, System, method and computer program for creating two dimensional (2D) or three dimensional (3D) computer animation from video.
Sampat Ketan (Portland OR) Kembel John (Lake Oswego OR), User interface, method, and apparatus selecting and playing channels having video, audio, and/or text streams.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.