自1950年代起，臺灣紡織工業發展出上中下游完整的生產體系，今全球主要大廠相繼投入AI技術發展，並以自身所擁有或累積的數據分析能力持續挖掘市場趨勢。對成衣製造業者而言，若能藉由自動化技術的應用，將對其勞動力密集的產業特性，提出節省人力與時間成本的改進。本研究應用關節辨識軟體OpenPose，偵測成衣作業員關節點作為現場績效的量測。結果顯示，在比較複雜的生產線環境，有不少情況相關的關節點的辨識都失效。尤其是在啟動與結束階段，其辨識率都非常低。反而在非啟動與非結束階段，或許是因為作業員的手腕不易為身體所遮掩，辨識的成功率提升許多。換言之，僅僅只是應用左/右手腕的關節點資訊來進行判讀還是不足。

Taiwan textile industry developed into a complete production system since 1950s. Major global manufacturers have successively invested in the development of AI technology and continue to tap with their own or accumulated data analysis capabilities. For garment manufacturers, the applications of AI, significant improvements to the labor-intensive industries will rise by saving labor and time costs. In this study, the skeleton analysis software OpenPose was adopted to detect the joint points of garment operators as a measure of on-site performance. The results show that in a relatively complex production line environment, there are many situations where the identification of related joint points fails. Especially in the start and end stages, the recognition rate is very low. On the contrary, in the non-starting and non-finishing phases, perhaps because the operator’s wrists are not hindered by body, the recognition rate is much higher. In sum, the use of the joints information of the left/right wrist for performance is not up to the need which would be improved with other joint information.

Chang, R.-Y., Pan, H.-Y., Lin, B.-L., Chen, W.-L., Hsieh, J. E., Huang, W.-Y., & Li, L.-H. (2018). AI Clerk: 會賣東西的機器人 (AI Clerk: Shopping Assistant)[In Chinese]. Paper presented at the Proceedings of the 30th Conference on Computational Linguistics and Speech Processing (ROCLING 2018).
Nunes, J. F., Moreira, P. M., & Tavares, J. M. R. (2016). Human motion analysis and simulation tools: a survey. In Handbook of Research on Computational Simulation and Modeling in Engineering (pp. 359-388): IGI Global.
Toshev, A., & Szegedy, C. (2014). Deeppose: Human pose estimation via deep neural networks. Paper presented at the Proceedings of the IEEE conference on computer vision and pattern recognition.
王琮文, 吳俊宇, 洪劭澂, 藍意萱, 羅瑞婷, 楊于萱, & 姜雅筑. (2018). 巡房機器人-貼心的智慧照服員. (學士), 中台科技大學,
田佳儒. (2016). 電子商務與大數據在壽險應用之研究. (碩士), 淡江大學,
光電新聞網-新聞. (2017). 微軟小冰發布新詩集，看看AI寫的詩歌味道如何？. Retrieved from https://read01.com/dozzK3.html#.XxBvROR7k2w.).Retrieved on.2020/5/27
余至浩. (2016). 製造業龍頭GE工廠明年導入AR 石油天然氣業務先採用. Retrieved from https://www.ithome.com.tw/news/109692.).Retrieved on.2020/5/30
林丽霞. (2009). 牛仔成衣后整理. 印染, 35(18), 29-34.
翁書婷. (2017). 癌症病患新選擇：台北醫學大學導入IBM華生人工智慧治療輔助系統. Retrieved from https://www.bnext.com.tw/article/45382/watson-for-oncology-ibm-cancer.).Retrieved on.2019/10/18
產業價質鏈資訊平台. (2019). 紡織產業鏈簡介. Retrieved from https://ic.tpex.org.tw/introduce.php?ic=O000.).Retrieved on.2019/October 25
陳詩涵. (2019). AI 機器人輔助情緒繪本教學對幼兒情緒能力之影響. (幼兒教育學系碩士班), 屏東大學,
勞動部. (2019). 成衣及服飾品製造業. Retrieved from https://www.mol.gov.tw/media/1662/ac03.pdf.).Retrieved on.2020/April 25
黃仲宏. (2018). 人工智慧技術將讓機器人如虎添翼. 機械工業雜誌(418), 24-31.
魏聰哲. (2019). 臺灣