Portugal

Being creative in movement-based improvisational environments, such as dance floors, poses a difficult challenge for computers. It is particularly challenging for computers to judge the creativity of inputs and responses in these open-ended domains. LuminAI is an improvisational dance installation where an artificially intelligent agent dances with a user. In this paper, we enable the agent to judge the creativity of dance gestures through the development of creativity metrics, including the novelty, value, and surprise of a gesture. We then use these metrics to implement a lead-and- follow dynamic: when the user is performing less creatively, the agent tries to lead by performing more creatively. A user study is performed to compare the original and lead-and-follow systems, with results showing users found the lead-and-follow system’s gestures lower in quality than the original, more surprising than the original, and found the original system more engaging and influential on their actions.

ICCC&#39;20

Creativity Metrics for a Lead-and-Follow Dynamic in an Improvisational Dance Agent

kinect

movement

improvisational

co-creative

lead-and-follow

turn-taking

creativity metrics

dance

Computational creativity is the art, science, philosophy and engineering of computational systems which, by taking on particular responsibilities, exhibit behaviours that unbiased observers would deem to be creative. As a field of research, this area is thriving, with progress in formalising what it means for software to be creative, along with many exciting and valuable applications of creative software in the sciences, the arts, literature, gaming and elsewhere.

The ICCC conference series organized by the Association for Computational Creativity since 2010 is the only scientific conference that focuses on computational creativity alone and also covers all aspects of it.


ICCC'20

The ICCC conference series organized by the Association for Computational Creativity since 2010 is the only scientific conference that focuses on computational creativity alone and also covers all aspects of it.
Computational creativity is the art, science, philosophy and engineering of computational systems which, by taking on particular responsibilities, exhibit behaviors that unbiased observers would deem to be creative. As a field of research, this area is thriving, with progress in formalizing what it means for software to be creative, along with many exciting and valuable applications of creative software in the sciences, the arts, literature, gaming and elsewhere.



poster

"This paper reports on the automatic selection of short-texts for amplifying the range of a given input's context, e.g. a news headline. Different methods are applied to select a corresponding expression from a list, which should be as semantically related to the input as possible. This study was developed for the Portuguese language, and considered expressions are proverbs, where figurative language is predominant.
The set of explored methods includes some based on word overlap, others on static word embeddings, and also on recent contextual embeddings. To compare the explored methods in this subjective scenario, a survey was answered by humans, who rated the value of the selected expressions in terms of relatedness with regard to the input, and the humoristic value that may arise from this selection. The main conclusion was that simpler approaches, which end up selecting expressions that share words with the headline, are more easily related and considered to be funnier than other more elaborate approaches, which are more focused on the context."

Comparing Different Methods for Assigning Portuguese Proverbs to News Headlines

Since the heydays of music informatics, around the 1950s, the modeling and prediction of musical structures manifested as symbolic representations have been continuously pursued. The operational property of such methods is to provide the conditional distribution over an alphabet -- i.e., the entire collection of unique musical events in a composition or corpus -- given a context -- i.e., a preceding sequence. This distribution unpacks temporal morphologies that support multiple applications for predictive and assisted creative tasks, such as the generation of new musical sequences that retain a structural resemblance to a modeled source. Despite their longstanding tradition, state-of-the-art methodologies for symbolic music modeling are yet to reach the music community. Naive models such as Markov chains, which are known to neglect the fundamental hierarchical nature of musical structure, remain common practice. In this paper, we extensively review existing methodologies for symbolic music representation and modeling, as the first steps towards a study on the resulting balance across familiarity and novelty in generative music applications.

Towards balanced tunes: A review of symbolic music representations and their hierarchical modeling

We describe a method and a proof of concept which allow the generation of rich and engaging dialogues between virtual characters from a formalised plot description. The structure of the dialogue generated borrows from inferential pragmatics, following the Geneva Model of discourse analysis, in order to provide realistic interaction between characters in the narrative. At a higher level, this discourse is organised following heuristics borrowed from narratology theory in order to elicit emotions linked to dramatic tension and thus favour narrative engagement. Besides enriching narrative generation systems embedded within simulation applications, our work also has the potential to be adapted to support engaging interactive dialogues between users and virtual conversational agents in narrative systems.

A Pragmatics-based Model for Narrative Dialogue Generation

"An interactive evolutionary system to generate letterings is presented. The system allows the creation of a wide range of alternative designs that can be used as stimuli for inspiration or for the creation of visual identities with different variations. This work began as a parametric system that generated glyph designs by recombining
parts of skeletons extracted from existing typefaces which are then filled with custom shapes. In this paper, we employ a Genetic Algorithm to evolve the input parameters of this parametric system. The experimental results show that the presented evolutionary system enables users to interactively create unique letterings according to their aesthetic preferences."

Which type is your type?

The importance of reproducibility has been recognised as one of the crucial elements of research especially in life sciences, but also in other disciplines, e.g. medicine, social sciences, natural language processing who have acknowledged the reproducibility crisis. In this paper, we discuss the issue of reproducibility in the field of computational creativity. We present the findings from an indirect reproducibility study that assesses the transparency of available information to allow reproducibility. The analysis of articles published in ICCC proceedings has shown that most computational creativity (CC) publications provide very little information with regard to documentation and resources needed for reproducibility and replicability of the published results. By reviewing best practices from the broader scientific community and considering the particularities of CC recommendations are put forward that will hopefully inspire the creation of practices related to reproducibility and improve the reusability of CC research within the community and beyond.

A Study on Reproducibility in Computational Creativity Research

We consider two works in computational criticism, a little studied but important area for computational creativity (Stiny and Gips 1978) & (Fisher and Shin 2019). The analysis results in the development of a more general, novel, extended model of computational criticism. Along with the more general model, we also provide the first formal attempt to discuss and distinguish meaningful computational criticism. This theoretical synthesis should be useful to the study and development of computational critics and help to spark further dialogue in this area.

Extending the Philosophy of Computational Criticism

Order and Progress: a sonic segue across A Auriverde is an abstract audiovisual sojourn across the Flag of Brazil realised as a sonified and politicised astronomy show, evoking emotions and responses through the use of fantasy, mystery, exaggeration, distortion and parallelism. The astronomy represented on the flag depicts a specific moment in time: the exact minute that Brazil declared its independence. Scientific data from these stars is treated as both raw material, a virtual generated stone sculpted into musical existence through figurative sonification; and as mathematical forces that forge the sounds and cultures of Brazil, both past and present, into an audiovisual blend of history, science, politics and ethnology.

Order and Progress: a sonic segue across A Auriverde

As a first step towards studying the ability of human crowds and machines to effectively co-create, we explore several human-only collaborative co-creation scenarios. The goal in each scenario is to create a digital sketch using a simple web interface. We find that settings in which multiple humans iteratively add strokes and vote on the best additions result in sketches with the highest creativity (combination of value and novelty). Lack of collaboration leads to a higher variance in quality and lower novelty or surprise. Collaboration without voting leads to high novelty but low quality.

Exploring Crowd Co-creation Scenarios for Sketches

## Please click on the link to watch the explanation video: https://screencast-o-matic.com/watch/cYjrolmlly



"Les Manifs - The Sound of Solidarity Is an audiovisual piece reflecting upon the artists’ experiences of the French ‘Gilets Jaunes’ protests in December 2018. 

Whilst living in the Occitanie region, the artists witnessed the rise of the protest movement sweeping the country. The grassroots group was sparked by rising fuel prices and the cost of living. The movement grew to encompass broader sentiments of domestic discontent and socioeconomic injustice. With wide support from many different demographics, weekly disruptive demonstrations expanded to almost every town in the country.

President Macron was quick to condemn the protests. He was categorical in his refusal to concede any of the movement’s demands. However the resolute tenacity of the Gilets Jaunes was underestimated. After 8 consecutive weeks of protests, Macron was forced to make concessions. Regardless, the protests continued unabated for 45 consecutive weeks in Paris, attracting over 300,000 demonstrators nationally at its height. 

As support for the movement grew, so too did the tension. Police clashed violently with rioters, known as the Ultra-Jaunes. Cities were heavily vandalised: rioters targeted banks, speed cameras, and parking meters. The Police response was equally controversial. Their aggressive tactics inflicted many (sometimes life changing) injuries to rioters and peaceful protestors alike.

The piece comprises binaural sound recordings and photographs of the 10th week of demonstrations in the city of Toulouse, where 10,000 protestors took to the streets. In gonzo-journalistic style, the photographer and recordist individually explored the protests through their mediums.

Both were drawn in with nightfall by the chaos and drama of the ultra-jaunes, where lawlessness reigns and revolution is romanticised.

The work portrays the story of a day of direct action with the Gilets Jaunes, following different groups and factions to a dangerous climax. Ultimately, the artists discover a place where for a fleeting moment, amidst barrages of tear gas, flash grenades and water cannons, some of the most disenfranchised members of society find their voices."



Les Manifs - The Sound of Solidarity

Light is one of the most enduring and resilient components of the Universe. The speed of light is a constant regardless of location. This video audio art work has been produced using two algorithmic regimes. 

The first regime involves scanning scientific experiment data visualisations, resulting in audio being generated from the visual data. The audio is then compiled into a soundtrack. The second regime involves putting the soundtrack into software that generates animated form. The production of the audio-video art work is cameraless. 

Light seen and unseen involved scanning scientific visualisations from various light wavelengths – at nonvisible (Gamma ray wavelength), UV, visible and infrared wavelengths. Visualisations used were of a photon shower from the Veritas atmospheric Cherenkov Telescope array in Arizona; a solar prominence eruption recorded on May 3rd 2013 at UV wavelengths; a Hubble image of the Serpens constellation of the Eagle Nebula giving birth to new stars, at visible light wavelengths; and an image of the nebula NCG2174 at infra-red wave-lengths.

Technical information
Light seen and unseen
Video mp4 encoding, 5 minutes 24 seconds
1920 x 1080 px, 29 frames per second
Stereo audio 48kHz 

Creativity Metrics for a Lead-and-Follow Dynamic in an Improvisational Dance Agent

Downloads

Next from ICCC'20

Comparing Different Methods for Assigning Portuguese Proverbs to News Headlines

Similar lecture

Towards Movement Generation with Audio Features