共用方式為


快速入門:開始對 Azure AI Studio 中的影像及影片使用含有視覺功能的 GPT-4 Turbo

重要

本文所述的部分功能可能僅適用於預覽版。 此預覽版本沒有服務等級協定,不建議將其用於生產工作負載。 可能不支援特定功能,或可能已經限制功能。 如需詳細資訊,請參閱 Microsoft Azure 預覽版增補使用條款

您可以閱讀本文,以開始使用 Azure AI Studio 來部署及測試含有視覺功能的 GPT-4 Turbo 模型。

含有視覺功能的 GPT-4 Turbo 和 Azure AI 視覺可提供進階功能,包括:

  • 光學字元辨識 (OCR):從影像擷取文字並與使用者的提示和影像結合,以展開內容。
  • 物件基礎:運用物件基礎彌補含有視覺功能的 GPT-4 Turbo 文字回應,並概述輸入影像中的主要物件。
  • 影片提示:含有視覺功能的 GPT-4 Turbo 藉由擷取與使用者提示最相關的影片畫面來回答問題。

使用 GPT-4 Turbo 搭配視覺和 Azure AI 視覺功能時,可能會收取額外的使用量費用。

必要條件

準備媒體

您需要影像來完成影像快速入門。 您可以使用此範例映像或任何其他可用的映像。

可用來完成快速入門的交通事故相片。

針對視訊提示,您需要長度低於三分鐘的視訊。

使用視覺模型部署 GPT-4 Turbo

  1. 登入 Azure AI Studio ,然後選取您想要運作的中樞。
  2. 在左側導覽功能表上,選取 [ AI 服務]。 選取 [ 試用 GPT-4 Turbo ] 面板。
  3. 在 gpt-4 頁面上,選取 [ 部署]。 在出現的視窗中,選取您的 Azure OpenAI 資源。 選取 vision-preview 作為模型版本。
  4. 選取部署
  5. 接下來,移至新模型的頁面,然後選取 [在遊樂場中開啟]。 在聊天遊樂場中,您應該在 [部署] 下拉式清單中選取您建立的 GPT-4 部署

在此聊天工作階段中,您會指示助理協助了解輸入的影像。

  1. 在 [系統訊息] 索引標籤上的 [系統訊息] 文字框中,提供此提示以引導助理:"You're an AI assistant that helps people find information."您可以針對影像或案例量身打造提示。
  2. 選取 [ 套用變更 ] 以儲存變更。
  3. 在聊天會話窗格中,選取 [附件] 按鈕,然後選取 [ 上傳影像]。 選擇您的映像。
  4. 在聊天欄位中新增下列問題: "Describe this image",然後選取要傳送的向右箭號圖示。
  5. 向右箭號圖示會取代為 [停止] 按鈕。 如果您選取它,助理會停止處理您的要求。 在這個快速入門中,請讓助理完成回覆。
  6. 助理會以影像的描述回復。
  7. 詢問與影像分析相關的後續問題。 您可以輸入 。 "What should I highlight about this image to my insurance company?"
  8. 您應該會收到類似此處所顯示內容的相關回應:
    When reporting the incident to your insurance company, you should highlight the following key points from the image:  
    
    1. **Location of Damage**: Clearly state that the front end of the car, particularly the driver's side, is damaged. Mention the crumpled hood, broken front bumper, and the damaged left headlight.  
    
    2. **Point of Impact**: Indicate that the car has collided with a guardrail, which may suggest that no other vehicles were involved in the accident.  
    
    3. **Condition of the Car**: Note that the damage seems to be concentrated on the front end, and there is no visible damage to the windshield or rear of the car from this perspective.  
    
    4. **License Plate Visibility**: Mention that the license plate is intact and can be used for identification purposes.  
    
    5. **Environment**: Report that the accident occurred near a roadside with a guardrail, possibly in a rural or semi-rural area, which might help in establishing the accident location and context.  
    
    6. **Other Observations**: If there were any other circumstances or details not visible in the image that may have contributed to the accident, such as weather conditions, road conditions, or any other relevant information, be sure to include those as well.  
    
    Remember to be factual and descriptive, avoiding speculation about the cause of the accident, as the insurance company will conduct its own investigation.
    

檢視和匯出程式碼

在聊天會話的任何時間點,您可以啟用 聊天視窗頂端的 [顯示原始 JSON] 參數,以查看格式化為 JSON 的交談。 以下是快速入門聊天工作階段開始時的樣子:

[
	{
		"role": "system",
		"content": [
			"You are an AI assistant that helps people find information."
		]
	},
]

清除資源

為了避免產生不必要的 Azure 費用,如果您不再需要在本快速入門中建立的資源,則應該加以刪除。 若要管理資源,您可以使用 Azure 入口網站

下一步