{"id":171165,"date":"2013-07-03T01:46:14","date_gmt":"2013-07-03T01:46:14","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/project\/rgb-d-dataset-7-scenes\/"},"modified":"2022-09-07T10:55:36","modified_gmt":"2022-09-07T17:55:36","slug":"rgb-d-dataset-7-scenes","status":"publish","type":"msr-project","link":"https:\/\/www.microsoft.com\/en-us\/research\/project\/rgb-d-dataset-7-scenes\/","title":{"rendered":"RGB-D Dataset 7-Scenes"},"content":{"rendered":"
\"RGB-D<\/a>The 7-Scenes dataset is a collection of tracked RGB-D camera frames. The dataset may be used for evaluation of methods for different applications such as dense tracking and mapping and relocalization techniques.<\/div>\n

<\/p>\n

\n

<\/h1>\n

<\/h2>\n

Overview<\/h2>\n

All scenes were recorded from a handheld Kinect RGB-D camera at 640×480 resolution. We use an implementation of the KinectFusion<\/a> system to obtain the ‘ground truth’ camera tracks, and a dense 3D model. Several sequences were recorded per scene by different users, and split into distinct training and testing sequence sets. Details on how this data can be used for example for the evaluation of relocalization methods can be found in our papers listed under publications.<\/p>\n

\"7-Scenes<\/span><\/span><\/p>\n

Data Description<\/h2>\n

For each scene, we provide one zip file which contains several sequences. Each sequence is a continuous stream of tracked RGB-D camera frames. Tracking\u00a0has been performed using ICP and frame-to-model alignment\u00a0with respect to a dense reconstruction\u00a0represented by a\u00a0truncated signed distance\u00a0volume.<\/p>\n

Each sequence (seq-XX.zip)\u00a0consists of 500-1000 frames. Each frame consists of three files:<\/p>\n