345 datasets found

XIMAGENET-12 is an explainable visual benchmark dataset for model robustness evaluation. It consists of over 200K images with 15,410 manual semantic annotations. The dataset is...

A dataset used for training a convolutional neural network (CNN) for object detection.

A dataset used for training a recurrent neural network (RNN) to infer positional uncertainties for a model predictive control (MPC) algorithm.

The AI-TOD remote sensing dataset is used for detecting dense small objects in aerial images.

The Visual7W dataset for Visual Question Answering (VQA). The dataset contains 7,000 images with 7,000 queries.

The ACE dataset is a large-scale dataset for image classification and object detection.

The INAT dataset is a large-scale dataset for image classification and object detection.

Object detection on Unmanned Aerial Vehicles (UAVs) is still a challenging task. The recordings are mostly sparse and contain only small objects.

The PointPillars dataset is a benchmark for object detection in 3D point clouds.

The Imagenette dataset used in the paper for class density and dataset quality in high-dimensional, unstructured data.

The Automotive Radar Dataset is a dataset containing data collected from a scanning Navtech radar in various weather conditions.

RADIATE is a radar dataset for automotive perception in bad weather, including 3 hours of annotated radar images with over 200K labelled road actors.

The UCAS-AOD dataset contains aerial images and annotated objects.

The DOTA, HRSC2016, and UCAS-AOD datasets are used for evaluating the proposed DARDet.

The Caltech dataset is a benchmark for pedestrian detection, consisting of approximately 10 hours of video taken from a vehicle driving through Los Angeles.

Scale-sensitive object detection remains a challenging task, where most of the existing methods could not learn it explicitly and are not robust to scale variance.

The dataset used in the paper is the COCO detection benchmark.

The FSCD-LVIS dataset is used to evaluate the proposed method.

Class-agnostic object counting aims to count all objects in an image with respect to example boxes or class names, a.k.a few-shot and zero-shot counting.

The COCO-Place Challenge 2017 is a benchmark for place recognition.