Flower (documentation video), 2016-17, Object recognition algorithm and custom software, Screen, projector and computer, Variable dimensions.

’Flower’ presents a series of distorted flower images which have still been recognised as "flower" by a computer vision*.

*This project uses the Label Detection of Google Cloud Vision API. The images are selected when "flower" is the first in the result label list and its score is higher than "0.9".

  • 01, Flower (selected image), 2016-17.

  • 01, Flower (selected image), 2016-17.

  • 01, Flower (part), 2016-17.


  • 02, Flower (part), 2016-17.


  • 03, Flower (part), 2016-17.


  • 04, Flower (part), 2016-17.


  • 05, Flower (part), 2016-17.


  • 06, Flower (part), 2016-17.


  • 07, Flower (part), 2016-17.


  • Flower (installation view at Gwangju Museum of Art, 2017), 2016-17, projection on 6 rear screens(3 at 400 x 225 cm and 3 at 480 x 270 cm).

  • Flower (installation view at Gwangju Museum of Art, 2017), 2016-17, projection on 6 rear screens(3 at 400 x 225 cm and 3 at 480 x 270 cm).

  • Flower (installation view at Gwangju Museum of Art, 2017), 2016-17, projection on 6 rear screens(3 at 400 x 225 cm and 3 at 480 x 270 cm).

  • Flower (installation view at Gwangju Museum of Art, 2017), 2016-17, projection on 6 rear screens(3 at 400 x 225 cm and 3 at 480 x 270 cm).

  • Flower (installation view at Pohang Museum of Steel Art, 2019), 2016-17, 7 channel video projection on the wall. Photo: Jihoon Byun.

  • Flower (installation view at Pohang Museum of Steel Art, 2019), 2016-17, 7 channel video projection on the wall. Photo: Jihoon Byun.