Choreography Synthesis Research Articles

Automated Synthesis of Service Choreographies

Future Internet research promotes the production of a distributed-computing environment that will be increasingly surrounded by a virtually infinite number of software services that can be composed to meet user needs. Services will be increasingly active entities that, communicating peer-to-peer, can proactively make decisions and autonomously perform tasks. Service choreography is a form of decentralized service composition that describes peer-to-peer message exchanges among participant services from a global perspective. In a distributed setting, obtaining the coordination logic required to realize a choreography is nontrivial and error prone. So, automatic support for realizing choreographies is needed. For this purpose, researchers developed a choreography synthesis tool. The Web extra at http://www.di.univaq.it/marco.autili/synthesis/shortdemo/demo.htm is a short demonstration of CHOReOSynt, a choreography synthesis tool.

Choreography Synthesis as Contract Agreement

We propose a formal model for distributed systems, where each participant advertises its requirements and obligations as behavioural contracts, and where multiparty sessions are started when a set of contracts allows to synthesise a choreography. Our framework is based on the CO2 calculus for contract-oriented computing, and borrows concepts and results from the session type literature. It supports sessions where the number of participants is not determined beforehand, and keeps CO2's ability to rule out participants that are culpable if contracts are not fulfilled at runtime. We show that we have progress and session fidelity in CO2, as a result of the honesty of participants - i.e., their ability to always adhere to their contracts.

Correction to "Learn2Dance: Learning Statistical Music-to-Dance Mappings for Choreography Synthesis" [Jun 12 747-759

In the above paper (ibid., vol. 14, no. 3, pp. 747-759, June 2012, p. 750), 44 was printed in error rather than 144. The corrected sentence and equation (6) are presented here.

Learn2Dance: Learning Statistical Music-to-Dance Mappings for Choreography Synthesis

We propose a novel framework for learning many-to-many statistical mappings from musical measures to dance figures towards generating plausible music-driven dance choreographies. We obtain music-to-dance mappings through use of four statistical models: 1) musical measure models, representing a many-to-one relation, each of which associates different melody patterns to a given dance figure via a hidden Markov model (HMM); 2) exchangeable figures model, which captures the diversity in a dance performance through a one-to-many relation, extracted by unsupervised clustering of musical measure segments based on melodic similarity; 3) figure transition model, which captures the intrinsic dependencies of dance figure sequences via an n-gram model; 4) dance figure models, capturing the variations in the way particular dance figures are performed, by modeling the motion trajectory of each dance figure via an HMM. Based on the first three of these statistical mappings, we define a discrete HMM and synthesize alternative dance figure sequences by employing a modified Viterbi algorithm. The motion parameters of the dance figures in the synthesized choreography are then computed using the dance figure models. Finally, the generated motion parameters are animated synchronously with the musical audio using a 3-D character model. Objective and subjective evaluation results demonstrate that the proposed framework is able to produce compelling music-driven choreographies.

An Approach to Automatic Motion Synthesis Harmonized with Music for Multiple Objects.

This paper proposed a research approach to automatic choreography synthesis based on SMF(Standard Midi File) for multiple animated figures. Based on the K.Hevner’s theory, 8 types of emotion in each beat can be extracted from music structure elements, such as tempo, key, rythm, melody, harmony, pitch, which can be computed from SMF. The time of one beat is limited to transmit the emotion to human. By the analysis of emotion, a music can be integrated to several time intervals, every which includes several continuous beats. The top value of synthetic emotion vector represents the emotion type of the interval. Based on the experiment result of C.Matsumoto, 14 motion factors can be mapped from the emotion of interval. According to the 14 motion factors, the macro-motions at the terminals of every interval, and the micromotions between every interval can be generated by mapping rules. We made a prototype system and did a subjective evaluation experiment. The result is fairly good at the congruity between generated motions and given emotion music. A successful research to solve these issues should lead to aid the designation of 3DCG animation.

Dancing with Music: An Approach to Choreography Synthesis Corresponding with Music Based on 3D Spring Model

This paper proposed an approach to automatic motion generation of 3D spring model corresponding with music. Music factors, which include basic information and Kansei information, could be extracted from MIDI message. According to the congruency of Kansei information between music and motion, several connection functions between motion and music were applied. Instead of designing animation in every scene, a sequence of motion-controllers parameters, which can finish a set of motions, are generated through a synthesis algorithm. A prototype system was developed for evaluation and it was confirmed that the appreciators were relatively satisfied with the generated animation based on music.