Actions (2023)
Perception engines techniques extended to video sequences. Each is an ink print created and recognized by neural networks trained to classify dynamic behaviour from videos of human actions. Shapes and colors in this sequence were generated by Microsoft's X-Clip models which process a stack of sequenced input images - extending CLIP for general video-language understanding. Each activity is chosen from Deepmind's Kinetics dataset which is the main evaluation metric for testing video understanding in machine learning. In each canvas print the motion is encoded using an op-art inspired style which is unrolled for evaluation as a video.
From the Actions series
20x16” screen print on canvas
2023 1/1
Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. The category "punching person (boxing)" taken from Deepmind's Kinetics dataset of high quality videos of human action classes.
Alternate digital version available here.
From the Actions series
20x16” screen print on canvas
2023 1/1
Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. The category "Kissing" taken from Deepmind's Kinetics dataset of high quality videos of human actions.
Alternate digital version available here.
From the Actions series
20x16” screen print on canvas
2023 1/1
Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. The category "petting cat" taken from Deepmind's Kinetics dataset of high quality videos of human actions.
Alternate digital version available here.
From the Actions series
20x16” screen print on canvas
2023 1/1
Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. The category "blowing nose" taken from Deepmind's Kinetics dataset of high quality videos of human actions.
Alternate digital version available here.
From the Actions series
20x16” screen print on canvas
2023 1/1
Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. The category "push up" taken from Deepmind's Kinetics dataset of high quality videos of human actions.
Alternate digital version available here.
From the Actions series
20x16” screen print on canvas
2023 1/1
Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. The category "counting money" taken from Deepmind's Kinetics dataset of high quality videos of human actions.
Alternate digital version available here.
From the Actions series
20x16” screen print on canvas
2023 1/1
Screen print created by neural networks trained to classify dynamic scenes from videos of human behavior. The repeating patterns can be unrolled into a looping video and recognized by computer vision systems. The category "climbing tree" taken from Deepmind's Kinetics dataset of high quality videos of human actions.
Alternate digital version available here.