³í¹®¸í |
°¡»óÇö½Ç ±â¹Ý 3Â÷¿ø °ø°£¿¡ ´ëÇÑ °¨Á¤ºÐ·ù µö·¯´× ¸ðµ¨ / Emotion Classification DNN Model for Virtual Reality based 3D Space |
ÀúÀÚ¸í |
¸íÁö¿¬(Myung, Jee-Yeon) ; ÀüÇÑÁ¾(Jun, Han-Jong) |
¼ö·Ï»çÇ× |
´ëÇѰÇÃàÇÐȸ³í¹®Áý °èȹ°è, Vol.36 No.04 (2020-04) |
ÆäÀÌÁö |
½ÃÀÛÆäÀÌÁö(41) ÃÑÆäÀÌÁö(9) |
ÁÖÁ¦¾î |
°¡»óÇö½Ç; °¨Á¤; ³úÆÄ; FFT; µö·¯´× ; Virtual Reality(VR); Emotion; Electroencephalography(EEG); Fast Fourier Transform(FFT); Deep Learning |
¿ä¾à1 |
º» ¿¬±¸ÀÇ ¸ñÀûÀº DNN (Deep Neural Networks) ¸ðµ¨À» »ç¿ëÇÏ¿© »ç¿ëÀÚÀÇ °¨Á¤, ƯÈ÷ VR (Virtual-Reality) ±â¹ÝÀÇ 3Â÷¿ø µðÀÚÀÎ ´ë¾È¿¡ ´ëÇÑ ³úÆÄ (EEG) ±â¹ÝÀÇ °¨Á¤À» ºÐ·ùÇÏ´Â °ÍÀÌ´Ù. »ç¿ëÀÚÀÇ °¨Á¤À» ÃøÁ¤Çϱâ À§ÇØ 4 °¡Áö À¯ÇüÀÇ VR °ø°£ÀÌ ±¸ÃàµÇ¾úÀ¸¸ç, °¢ Àڱؿ¡ ´ëÇÑ ³úÆÄ°¡ ÃøÁ¤µÇ¾ú´Ù. EEG µ¥ÀÌÅÍ¿¡ ±âÃÊÇÑ Á¤·®Àû Æò°¡¿¡ ´õÇÏ¿©, VR ÀÚ±Ø »çÀÌÀÇ Â÷À̰¡ ÀÖ´ÂÁö¸¦ Á¤¼ºÀûÀ¸·Î È®ÀÎÇϱâ À§ÇÑ ¼³¹®ÀÌ ¼öÇàµÇ¾ú´Ù. Á¤±ÔÈ ¼øÀ§ ºÐ¼® °á°ú °èȹ À¯Çü °£¿¡ À¯ÀÇ ÇÑ Â÷À̰¡ È®ÀεǾú´Ù. µû¶ó¼ ÁÖ°üÀû ¼³¹®ÁöÀÇ °ªÀ» DNN ¸ðµ¨ÀÇ ¶óº§¸µ µ¥ÀÌÅÍ·Î, ¼öÁýµÈ EEG µ¥ÀÌÅ͸¦ ¸ðµ¨ÀÇ Æ¯Â¡ °ªÀ¸·Î »ç¿ëÇß´Ù.??¸ðµ¨ ±¸Ãà ¹× ÈÆ·Ã¿¡´Â Google Tensor Flow¸¦ »ç¿ëÇß´Ù. °á°úÀûÀ¸·Î °³¹ßµÈ ¸ðµ¨ÀÇ Á¤È®µµ´Â 98.9 %·Î ÀÌÀü ¿¬±¸º¸´Ù ³ô´Ù. µû¶ó¼ º» ¿¬±¸¿¡¼ Á¦¾ÈÇÑ ¸ðµ¨À» Ȱ¿ëÇÏ¿© VR ±â¹Ý 3Â÷¿ø ¼³°è ´ë¾È¿¡ ´ëÇÑ ¿¹ºñ»ç¿ëÀÚÀÇ °¨Á¤ÆÄ¾ÇÀÌ °¡´ÉÇØÁú °ÍÀ¸·Î ±â´ëµÈ´Ù. |
¿ä¾à2 |
The purpose of this study was to investigate the use of the Deep Neural Networks(DNN) model to classify user¡¯s emotions, in particular Electroencephalography(EEG) toward Virtual-Reality(VR) based 3D design alternatives. Four different types of VR Space were constructed to measure a user¡¯s emotion and EEG was measured for each stimulus. In addition to the quantitative evaluation based on EEG data, a questionnaire was conducted to qualitatively check whether there is a difference between VR stimuli. As a result, there is a significant difference between plan types according to the normalized ranking method. Therefore, the value of the subjective questionnaire was used as labeling data and collected EEG data was used for a feature value in the DNN model. Google TensorFlow was used to build and train the model. The accuracy of the developed model was 98.9%, which is higher than in previous studies. This indicates that there is a possibility of VR and Fast Fourier Transform(FFT) processing would affect the accuracy of the model, which means that it is possible to classify a user¡¯s emotions toward VR based 3D design alternatives by measuring the EEG with this model. |