Search Results for "dataset"

...We aim to build a tool that can be used for benchmarking SOTA models, while also allowing practitioners to efficiently pursue research into point cloud analysis, with the end goal of building models which can be applied to real-life applications. Task driven implementation with dynamic model and dataset resolution from arguments. Core implementation of common components for point cloud deep learning - greatly simplifying the creation of new models. 4 Base Convolution base classes to simplify the implementation of new convolutions. Each base class supports a different data format.

The dataset consists of 15K annotated video clips supplemented with over 4M annotated images in the following categories: bikes, books, bottles, cameras, cereal boxes, chairs, cups, laptops, and shoes. In addition, to ensure geo-diversity, our dataset is collected from 10 countries across five continents. Along with the dataset, we are also sharing a 3D object detection solution for four categories of objects — shoes, chairs, mugs, and cameras.

PhotoCloud is a tool to browse and a navigate for datasets composed of (a) a collection of registered images and (b) a 3D model over which images are aligned. The 3D model can be a point-cloud, triangle mesh. PhotoCloud is based on a client/server architecture and it is scalable: very large dataset (both in terms of (a) and (b) can be handled.