作業幫直播課如何用AI實現專注力和離席檢測

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"2020年1月起，專注力檢測、離席檢測、距離檢測功能先後在作業幫直播課產品“小組直播間”內上線。藉助AI視覺識別技術，研發團隊滿足輔導老師、家長、學生的切實需求，解決了孩子上網課“溜號”、“距離屏幕過近”等實際問題，讓人工智能技術於在線教育場景中真正落地。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"AI教具：爲輔導老師、家長減負"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"保持學生課堂專注力是提高課堂效率的有效途徑之一。在傳統雙師制線上大班直播課，輔導老師是課堂專注力的“人工監督員”，同時負責一兩百名學員，很難照顧到每一個人。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在作業幫“小組直播間”，課堂，輔導老師則裝備上了高效智能的“AI教具”。基於AI視覺識別技術，研發團隊開發出專注力檢測和離席檢測功能。當系統依據學生上課表現識別出專注力較低的學生時，跟課的輔導老師將會收到提示，並可一鍵點開學生的直播視頻查看，然後根據實際情況決定是否需要提醒學生。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/0d\/bb\/0ddc6567fb86ea65227b3b8fbb5ee0bb.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在2020年底的一次調研中，87.8%的輔導老師認爲，該功能對工作很有幫助，不僅在上課時，能提醒走神的孩子認真聽課，在課後的回訪中，也可以向家長反映孩子的聽課情況。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"現在，對於輔導老師來說，專注力和離席檢測已經成爲高效的輔助“教具”，在學生“溜號”走神的時候及時把他拉回來，而當檢測到學生已經離席時，輔導老師可能就要直接“找家長”了。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在系列功能設計時，王巖團隊瞭解到，很多家長都在給孩子報網課後想了一些方法，讓孩子與屏幕和書桌保持一定距離，“家長們對於小朋友的視力問題很在意。”"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"敏銳捕捉到家長這一“剛需”，團隊研發出距離檢測功能。通過AI視覺識別技術，系統可以判斷孩子眼睛與屏幕的距離，每隔15分鐘左右提醒一次。一旦屏幕距離小於規定參數，學生端便會彈出文案警告，直至坐姿矯正後，警告纔會消失。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"AI研發：“有用”大於“炫酷”"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在王巖看來，無論是距離檢測還是專注力、離席檢測，這一系列依靠AI技術的視覺互動功能背後，都有着共通的研發邏輯：定義任務，收集樣本。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在定義問題上，“眼睛與屏幕的距離”以及“學生是否在畫面中”都可以較爲輕鬆地判斷，但團隊對於“專注”這一較爲主觀的因素，卻很難給予一個清晰的定義標準。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"無論在哪種場景下，專注力都很難有標準定義，“它的維度太多了，比如有的學生會習慣性的偏頭，我可能認爲他不是很專注，或者是手上總有小動作，比如學生會經常摳手，這都是一些很主觀的判斷。”"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"從技術層面上來講，定義專注也是件難事。王巖在研究學生聽課狀態時發現，有的學生可能在發呆，但是畫面呈現出來的和沉思沒有什麼太大區別。這就對AI視覺分析識別提出更高的要求。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在多次討論和權衡之後，研發團隊將頭部、手部、肢體、表情等多種行爲結合制定出一套判斷標準。“不能說100%，但是準確率極大提高。”王巖表示。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"對於技術團隊來說，研發難度並非是決定是否要研發某一功能的關鍵，對教育是否有實際用途纔是決定因素，“我們設計這些功能都是從功用出發，希望它們在學習中能夠產生實際價值，而非炫酷。”"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"AI應用：在線教育爲人工智能落地提供良好契機"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"“智能機器人”、“無人駕駛”可能是大多數人對於AI技術的印象，但在王巖看來，人工智能最廣泛的落地場景，是在線教育領域。“AI技術在在線教育領域的應用雖然沒有那麼高深，但它的使用者卻足夠多。那麼多用戶，我們的一個小功能，很多人都能使用到。”"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"艾媒諮詢發佈的《2020中國K12在線教育行業研究報告》指出，人工智能技術已經成"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"爲K12在線教育的重要組成部分。目前，人工智能技術已被運用於在線教育課程、服務和教學管理的全流程。"}]}]}