다음을 통해 공유


빠른 시작: Azure AI Studio의 이미지 및 비디오에서 GPT-4 Turbo with Vision 사용을 시작합니다.

Important

이 문서에 설명된 기능 중 일부는 미리 보기로만 제공될 수 있습니다. 이 미리 보기는 서비스 수준 계약 없이 제공되며, 프로덕션 워크로드에는 권장되지 않습니다. 특정 기능이 지원되지 않거나 기능이 제한될 수 있습니다. 자세한 내용은 Microsoft Azure Preview에 대한 추가 사용 약관을 참조하세요.

이 문서를 통해 Azure AI Studio 사용을 시작하여 GPT-4 Turbo with Vision 모델을 배포하고 테스트합니다.

GPT-4 Turbo with Vision 및 Azure AI Vision은 다음과 같은 고급 기능을 제공합니다.

  • OCR(광학 문자 인식): 이미지에서 텍스트를 추출하고 이를 사용자의 프롬프트 및 이미지와 결합하여 컨텍스트를 확장합니다.
  • 개체 그라운딩: 개체 그라운딩으로 GPT-4 Turbo with Vision 텍스트 응답을 보완하고 입력 이미지에서 두드러진 개체를 윤곽선으로 표시합니다.
  • 비디오 프롬프트: GPT-4 Turbo with Vision은 사용자의 프롬프트와 가장 관련된 비디오 프레임을 검색하여 질문에 대답할 수 있습니다.

Vision 및 Azure AI Vision 기능과 함께 GPT-4 Turbo를 사용하는 경우 추가 사용 요금이 적용될 수 있습니다.

필수 조건

미디어 준비

이미지 빠른 시작을 완료하려면 이미지가 필요합니다. 이 샘플 이미지 또는 사용 가능한 다른 이미지를 사용할 수 있습니다.

빠른 시작을 완료하는 데 사용할 수 있는 자동차 사고 사진입니다.

비디오 프롬프트의 경우 길이가 3분 미만인 비디오가 필요합니다.

GPT-4 Turbo with Vision 모델 배포

  1. Azure AI 스튜디오에 로그인하고 작업하려는 허브를 선택합니다.
  2. 왼쪽 탐색 메뉴에서 AI 서비스를 선택합니다. GPT-4 Turbo 사용해 보기 패널을 선택합니다.
  3. gpt-4 페이지에서 배포를 선택합니다. 표시되는 창에서 Azure OpenAI 리소스를 선택합니다. 모델 버전으로 vision-preview를 선택합니다.
  4. 배포를 선택합니다.
  5. 다음으로, 새 모델의 페이지로 이동하여 플레이그라운드에서 열기를 선택합니다. 채팅 플레이그라운드에서 자신이 만든 GPT-4 배포는 배포 드롭다운에서 선택해야 합니다.

이 채팅 세션에서는 입력한 이미지를 이해하는 데 도움을 주도록 도우미에게 지시합니다.

  1. 시스템 메시지 탭의 시스템 메시지 텍스트 상자에서 도우미를 안내하는 이 프롬프트를 제공합니다. "You're an AI assistant that helps people find information." 프롬프트를 이미지 또는 시나리오에 맞게 조정할 수 있습니다.
  2. 변경 내용 적용을 선택하여 변경 내용을 저장합니다.
  3. 채팅 세션 창에서 첨부 파일 단추를 선택한 다음 이미지를 업로드합니다. 이미지를 선택합니다.
  4. 채팅 필드에 "Describe this image"라는 질문을 추가한 다음, 오른쪽 화살표 아이콘을 선택하여 보냅니다.
  5. 오른쪽 화살표 아이콘이 중지 단추로 바뀝니다. 이 옵션을 선택하면 도우미가 요청 처리를 중지합니다. 이 빠른 시작에서는 도우미가 회신을 완료하도록 합니다.
  6. 도우미는 이미지에 대한 설명과 함께 회신합니다.
  7. 이미지 분석과 관련된 후속 질문을 합니다. "What should I highlight about this image to my insurance company?"라고 입력할 수 있습니다.
  8. 여기에 표시된 내용과 유사한 관련 응답을 받아야 합니다.
    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.
    

코드 보기 및 내보내기

채팅 세션의 어느 지점에서든 채팅 창 맨 위에 있는 원시 JSON 표시 스위치를 활성화하여 JSON으로 형식이 지정된 대화를 볼 수 있습니다. 빠른 시작 채팅 세션의 시작 부분에서의 모습은 다음과 같습니다.

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

리소스 정리

불필요한 Azure 비용이 발생하지 않도록 하려면 이 빠른 시작에서 만든 리소스가 더 이상 필요하지 않은 경우 삭제해야 합니다. 리소스를 관리하려면 Azure Portal을 사용하면 됩니다.

다음 단계