SESHADRI SASIKALA - Academia.edu (original) (raw)
Uploads
Papers by SESHADRI SASIKALA
• Holistic scene understanding is a major goal in recent research of computer vision. • To deal... more • Holistic scene understanding is a major goal in recent research of computer vision.
• To deal with this task, reasoning the 3D relationship of components in a scene is identified as one of the key problems. The problem of estimating detailed 3D structure from a single image of an unstructured environment is considered.
• The goal is to create 3D models which are both quantitatively accurate as well as visually pleasing. Consequently, scene categorization is identified as the first step towards robust and efficient depth estimation from single images. Stage information serves as the first approximation of global depth, narrowing down the search space in depth estimation and object localization. Classification results demonstrate that stages can be efficiently learned from low-dimensional image representations.
• An extension method which gives better results which is based on point cloud concept is proposed. Here the first and second images of a single image are considered and 3D layout. is constructed. Experiments show that the results of our model outperform the state-of-the-art methods for 3D structure classification
• Holistic scene understanding is a major goal in recent research of computer vision. • To deal... more • Holistic scene understanding is a major goal in recent research of computer vision.
• To deal with this task, reasoning the 3D relationship of components in a scene is identified as one of the key problems. The problem of estimating detailed 3D structure from a single image of an unstructured environment is considered.
• The goal is to create 3D models which are both quantitatively accurate as well as visually pleasing. Consequently, scene categorization is identified as the first step towards robust and efficient depth estimation from single images. Stage information serves as the first approximation of global depth, narrowing down the search space in depth estimation and object localization. Classification results demonstrate that stages can be efficiently learned from low-dimensional image representations.
• An extension method which gives better results which is based on point cloud concept is proposed. Here the first and second images of a single image are considered and 3D layout. is constructed. Experiments show that the results of our model outperform the state-of-the-art methods for 3D structure classification