US 11,995,861 B2
Information processing system, content generation device, content presentation device, content generation method, content presentation method, and program
Kazu Miyakawa, Musashino (JP); Shingo Kinoshita, Musashino (JP); Kenya Suzuki, Musashino (JP); Satoshi Sakuma, Musashino (JP); and Tatsuya Matsui, Musashino (JP)
Assigned to NIPPON TELEGRAPH AND TELEPHONE CORPORATION, Tokyo (JP)
Appl. No. 17/594,028
Filed by NIPPON TELEGRAPH AND TELEPHONE CORPORATION, Tokyo (JP)
PCT Filed Mar. 19, 2020, PCT No. PCT/JP2020/012447
§ 371(c)(1), (2) Date Sep. 30, 2021,
PCT Pub. No. WO2020/203368, PCT Pub. Date Oct. 8, 2020.
Claims priority of application No. 2019-070141 (JP), filed on Apr. 1, 2019.
Prior Publication US 2022/0148215 A1, May 12, 2022
Int. Cl. G06T 7/70 (2017.01); G06V 10/40 (2022.01)
CPC G06T 7/70 (2017.01) [G06V 10/40 (2022.01); G06T 2207/10016 (2013.01)] 6 Claims
OG exemplary drawing
 
1. An information processing system comprising a content generation device configured to generate content information relating to presentation of content, and a content presentation device configured to control presentation of the content,
wherein
the content generation device includes at least one processor configured to
acquire an input video image,
detect position information of an object from the input video image,
extract attribute information of the object detected from the input video image acquired, and
generate the content information corresponding to the position information detected and the attribute information extracted, the content information including information to display a wave based on the object detected, and
the content presentation device includes at least one processor configured to
control the presentation of the content according to the content information generated,
wherein
the object detected includes a plurality of people,
the attribute information includes gender, age, height, clothes, and motion of the plurality of people,
a level of undulation of the wave depends upon the gender of the plurality of people,
the level of undulation is softer when the gender indicates more female people in the plurality of people than male people than when the gender indicate more male people in the plurality of people than female people,
a shape of the wave is based on a probability density function generated based on a Gaussian mixture model,
the probability density function is given by:
p(x═π,μ,M)=ΣkπkN(x═μk,Mk)
where π is a mixing coefficient, μ is a mean vector, M is a covariance matrix, N is a normal distribution, and K is a cluster number,
K is based on a total number of areas of the input video image in which the object is detected, and
the wave is generated based on the probability density function, a smoothing function based on the age, a smoothing function based on the height, and the level of undulation.