RISS 학술연구정보서비스

검색
다국어 입력

http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

변환된 중국어를 복사하여 사용하시면 됩니다.

예시)
  • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
  • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
닫기
    인기검색어 순위 펼치기

    RISS 인기검색어

      악천후 환경에서의 깊이 추정을 위한 적응형 이중 신뢰도 기반 카메라-레이더 융합 기법 = Adaptive Dual-Confidence Fusion for Camera-Radar Depth Estimation under Adverse Weather Conditions

      한글로보기

      https://www.riss.kr/link?id=T17413478

      • 0

        상세조회
      • 0

        다운로드
      서지정보 열기
      • 내보내기
      • 내책장담기
      • 공유하기
      • 오류접수

      부가정보

      다국어 초록 (Multilingual Abstract) kakao i 다국어 번역

      Under adverse weather conditions, camera-based depth estimation often becomes unreliable due to reduced contrast and severe visual degradation. Radar measurements remain stable under these conditions, but they are extremely sparse and provide limited structural cues. Because the reliability of the two sensors changes differently depending on the environment, existing fusion methods often fail to maintain consistent depth scales, especially at long ranges.
      To address this issue, this work proposes Adaptive Dual-Confidence Fusion Network(ADCNet), a camera-radar fusion model that explicitly estimates and calibrates the confidence of each sensor before fusion. ADCNet first predicts camera and radar confidences separately to capture reliability variations under rain or snow. The predicted confidences are then calibrated using temperature scaling to mitigate over-confident predictions. Finally, the calibrated confidences are normalized with a softmax function and used to adaptively fuse the two sensor features.
      We evaluate ADCNet on the Boreas dataset, which includes a wide range of adverse-weather scenes. Compared with AFNet and existing camera-radar fusion methods(Li et al., CaFNet), ADCNet consistently reduces MAE, RMSE, iMAE, iRMSE, and AbsRel. Qualitative results further demonstrate that ADCNet produces more stable and visually consistent depth maps in heavy rain and snow than existing methods.
      Overall, the results indicate that modeling sensor reliability and adaptively fusion camera and radar improves robustness under challenging weather, and that even sparse radar signals can stabilize depth estimation when integrated appropriately.
      번역하기

      Under adverse weather conditions, camera-based depth estimation often becomes unreliable due to reduced contrast and severe visual degradation. Radar measurements remain stable under these conditions, but they are extremely sparse and provide limited ...

      Under adverse weather conditions, camera-based depth estimation often becomes unreliable due to reduced contrast and severe visual degradation. Radar measurements remain stable under these conditions, but they are extremely sparse and provide limited structural cues. Because the reliability of the two sensors changes differently depending on the environment, existing fusion methods often fail to maintain consistent depth scales, especially at long ranges.
      To address this issue, this work proposes Adaptive Dual-Confidence Fusion Network(ADCNet), a camera-radar fusion model that explicitly estimates and calibrates the confidence of each sensor before fusion. ADCNet first predicts camera and radar confidences separately to capture reliability variations under rain or snow. The predicted confidences are then calibrated using temperature scaling to mitigate over-confident predictions. Finally, the calibrated confidences are normalized with a softmax function and used to adaptively fuse the two sensor features.
      We evaluate ADCNet on the Boreas dataset, which includes a wide range of adverse-weather scenes. Compared with AFNet and existing camera-radar fusion methods(Li et al., CaFNet), ADCNet consistently reduces MAE, RMSE, iMAE, iRMSE, and AbsRel. Qualitative results further demonstrate that ADCNet produces more stable and visually consistent depth maps in heavy rain and snow than existing methods.
      Overall, the results indicate that modeling sensor reliability and adaptively fusion camera and radar improves robustness under challenging weather, and that even sparse radar signals can stabilize depth estimation when integrated appropriately.

      더보기

      목차 (Table of Contents)

      • 1. 서론 1
      • 1.1 연구 배경 1
      • 1.2 연구 목적 4
      • 1.3 논문 구성 7
      • 2. 배경지식 및 관련 연구 8
      • 1. 서론 1
      • 1.1 연구 배경 1
      • 1.2 연구 목적 4
      • 1.3 논문 구성 7
      • 2. 배경지식 및 관련 연구 8
      • 2.1 자율주행 시스템 8
      • 2.2 깊이 추정 11
      • 2.3 카메라-레이더 융합 깊이 추정 13
      • 2.4 딥러닝 기반 불확실성 및 신뢰도 추정 16
      • 3. ADCNet: 적응형 이중 신뢰도 기반 카메라-레이더 융합 기법19
      • 3.1 ADCNet 아키텍처 21
      • 3.2 센서별 특성을 고려한 독립적 신뢰도 추정 23
      • 3.3 신뢰도 기반의 적응형 융합 메커니즘29
      • 3.4 불확실성 학습을 위한 손실 함수 설계 33
      • 4. 실험 및 성능 평가 37
      • 4.1 실험 환경 37
      • 4.2 실험 결과 40
      • 4.2.1 전체 성능 비교 40
      • 4.2.2 거리 구간별 성능 분석 44
      • 4.2.3 거리 증가에 따른 스케일 일관성과 이중 신뢰도 기반 불확실성 규제 분석 48
      • 4.2.4 악천후 환경에서의 시각적 비교52
      • 4.2.5 맑은 환경에서의 시각적 비교 55
      • 5. 결론 57
      • 참고문헌 59
      • 영문요약 67
      더보기

      분석정보

      View

      상세정보조회

      0

      Usage

      원문다운로드

      0

      대출신청

      0

      복사신청

      0

      EDDS신청

      0

      동일 주제 내 활용도 TOP

      더보기

      주제

      연도별 연구동향

      연도별 활용동향

      연관논문

      연구자 네트워크맵

      공동연구자 (7)

      유사연구자 (20) 활용도상위20명

      이 자료와 함께 이용한 RISS 자료

      나만을 위한 추천자료

      해외이동버튼