作品名稱:『 Hijacker:{}


媒材:EEG cap, EEG sensors, f/NIRS sensor, customize software, dataset, 

尺寸:尺寸可變,裱框尺寸 50cm * 50cm



Hijacker:{}〉為2010年與荷蘭UMGC醫學中心(University of Medical Center Groningen)合作,觀察早產嬰兒腦部血液活動狀態而發想的創作。作品以「夢的照相機」為主要概念,運用機器學習(machine learning)重新轉換2010年因資料量判讀與其中結構高度複雜性而無法呈現的部分延伸創造更多可能性。作品藉由賦予軟體程式特定的規則,讓腦波偵測機械真正具有「想像」的能力,藉由合成「夢境圖像」、做夢者腦波所產生的關鍵字資料,與做夢者記憶相互比較,讓第三方得以進一步探究他人的夢境,達到替做夢者想像出可能、或相似的夢境場景。


Title:『 Hijacker:{}


MaterialEEG cap, EEG sensors, f/NIRS sensor, customize software, dataset

Sizesize variable, frame size 50(H)cm *50(w)cm


"We can only use our imagination to understand and get an outline of people's dreams”


"Hijacker: {,}" is a concept that was conceived when we tested and monitored the blood activity of the brain of the early-born baby in cooperation with UMGC (University of Medical Center Groningen) in 2010.

At that time, because some issue and technical difficulty of designing data structure were more complicated. So this time, applying the way of Machine learning was recombined into this work, and redesign and simplify the flow of collecting data become easier, more flexible and create more possibilities. 

By assigning certain rules to the software, the work can imagine possible or similar dream scenes for dreamers. In a symbolic way to create a way of photography of dream as the main concept to interpret the imagination that is given to the machine.

By synthesizing the dream image, the keywords generated by the software based on the dataset of the brain activities are compared with the text description of the dreamer.

Basic procedure

Dataset (Daily){set and category by Image and Keyword} ->

Dataset (sleep){collecting base on EEG & NIRS} ->

Compare and re-category by Machine learning -> pick up keywords <-> compose sentence by ML <-> parsing images and train new dataset<-> get image composition by cocoapi -> generate image.