A Multiple Receptive Field Anti-occlusion Self-supervision Network with CBAM for Light Field Angular Super-resolution
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
By recording high-dimensional light data, a light field (LF) can accurately perceive a dynamic environment,thus supporting the understanding and decision-making of intelligent systems. However, with the discretesampling of high-dimensional signals, LF faces have insufficient efficacious acquisition of LF information. This study tackles this problem by introducing a self-supervised learning approach that uses convolutionalneural networks with varying receptive fields for processing sparse view inputs and subsequently generating adense view through warping. The primary basis relies on the fact that the inherent correlation of the LF dataand the convolutional block attention module (CBAM) are applied to process the LF data and wrap theoperation into a layer to construct a deep network. The proposed method eliminates occlusions and achievessuper-resolution LF angle reconstruction. Extensive experiments on an HCI dataset demonstrated that theproposed model outperforms recent state-of-the-art models.
Abstract
By recording high-dimensional light data, a light field (LF) can accurately perceive a dynamic environment,thus supporting the understanding and decision-making of intelligent systems. However, with the discretesampling of high-dimensional signals, LF faces have insufficient efficacious acquisition of LF information. This study tackles this problem by introducing a self-supervised learning approach that uses convolutionalneural networks with varying receptive fields for processing sparse view inputs and subsequently generating adense view through warping. The primary basis relies on the fact that the inherent correlation of the LF dataand the convolutional block attention module (CBAM) are applied to process the LF data and wrap theoperation into a layer to construct a deep network. The proposed method eliminates occlusions and achievessuper-resolution LF angle reconstruction. Extensive experiments on an HCI dataset demonstrated that theproposed model outperforms recent state-of-the-art models.