The perceptron Lab was founded by Héctor Rodríguez. Its main aim is to use machine learning and computer vision methods to analyze film aesthetics automatically and to visualize those analyses in the form of video installations. The main research direction to this point has focused mainly on the application of optical flow, matrix factorization, and deep neural networks to the visualization of cinematic movement.

 

PROJECTS

Hidden Networks

Hidden Networks is a large-scale eight-channel video installation that applies machine learning techniques to the analysis of the moving image. A system of deep neural networks analyzes the optical flow in a dataset of silent films directed by Louis Feuillade and identifies scenes with similar motions: scenes in which the figures move in the same direction, with the same speed, or with the same rhythm. The work was selected by a jury convened by the government of the Canary Islands for presentation at the El Tanque Cultural Centre in Tenerife. 

A video documentation of the installation can be found at https://vimeo.com/581860190

The following website introduces the concept of the work:
http://concept-script.com/hidden_networks/index.html#

 

Image gallery 

3D model used in the production of the work, these images are the actual installation in the El Tanque Cultural Centre. 

Images in the El Tanque Cultural Centre

 

Images in the El Tanque Cultural Centre

 

Images in the El Tanque Cultural Centre

 

Images in the El Tanque Cultural Centre

 

Images in the El Tanque Cultural Centre

 

Images in the El Tanque Cultural Centre

 

Images in the El Tanque Cultural Centre

 

Recent Exhibition 

Héctor Rodríguez, Hidden Networks (solo exhibition), El Tanque Cultural Center (Tenerife, Spain), 3 July, 2021. 

For more information about the exhibition, please visit 
http://www.gobiernodecanarias.org/cultura/eltanque/eventos/RedesOcultas

 

Errant version 2: The Kinetic Propensity of Images

A project about the automatic analysis and visualization of motion in the cinema. A newly designed machine learning algorithm decomposes the movement in every sequence of a movie into a set of elementary motions. These elementary motions are then recombined to produce a reconstruction of the visible movement in the sequence. The analysis and reconstruction are displayed as a two-channel video installation. The visualization of the movement uses a variant of the streakline method often employed in fluid dynamics. 

Shown at: Neural Information Processing Systems NEURLPS, December 9, 2020.

http://www.aiartonline.com/highlights-2020/hector-rodriguez-3/

 

Deep Archetypes: Meditation on Violence

This video uses deep learning and archetypal analysis methods to analyze and visualize the rhythmic flow of Maya Deren's 1948 film Meditation on Violence, made in collaboration with Chinese martial artist Chao-Li Chi (Ji Chaoli).

Shown at: Conference on Computer Vision and Pattern Recognition (CVPR) 2021: Computer Vision Art Gallery, 19 June 2021

https://computervisionart.com/pieces2021/deep-archetypes/

<drupal-entity data-embed-button="media_browser" data-entity-embed-display="view_mode:media.embedded" data-entity-type="media" data-entity-uuid="489e51ab-aa3f-4399-adfe-cbd94c3149aa" data-langcode="en"></drupal-entity>
2020

Exhibition grant of the Canary Island Institute for Cultural Development and the Government of the Canary Islands, Spain.

This grant covered the cost of producing an exhibition in the El Tanque Cultural Space in Tenerife, Spain. 

2021

[Solo Exhibition] Redes Ocultas: Imagen en Movimiento y Visión Artificial  (Hidden Networks: Computer Vision and Moving Image), 
Venue: El Tanque Cultural Space, Spain, 2021/06/29 - 2021/09/25 

URL: http://www.gobiernodecanarias.org/cultura/eltanque/eventos/RedesOcultas

Hidden Networks is a large-scale, eight-channel video installation that uses deep learning methods to analyze and compare movement in the films of French director Louis Feuillade. This work was the basis of a major solo exhibition in a large space (50 meters in diameter, 20 meters in height). It was subsidized in part with a grant from the government of the Canary Islands.

2019

“Algorithmic Analysis and Visualization of Motion in Cinema”,  12th ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia, SIGGRAPH Asia 2019, Brisbane, Australia, 17/11/19 - 20/11/19. 

http://sa2019.conference-program.com/presentation/?id=gp_197&sess=sess138

This presentation explores some of the possible applications of unsupervised machine learning methods in found footage cinema, a tradition of experimental art that re-edits excerpts from existing films. This artistic practice sometimes aims to reconfigure our experience of the moving image heritage. In this context, machine learning algorithms has the potential to capture aspects of the cinematic experience for which we lack critical concepts, and which are for this reason difficult to describe. One important example concerns cinematic motion. Established critical discourse often speaks of motion in film by reference to the movement of objects or the camera. Film scholars might describe a scene by noting, for instance, that a person is walking fast or that the camera is tilting upwards. What is missing in this kind of description is the visual texture of cinematic movement.

People

 
Associate Professor
RODRIGUEZ, Hector
羅海德
VIEW PROFILE