1,082 datasets found

The conventional lottery ticket hypothesis (LTH) claims that there exists a sparse subnetwork within a dense neural network and a proper random initial-ization method called the...

Internal head and neck CT scans for normal organ segmentation

Head and neck CT scans for normal organ segmentation

Vision transformers have achieved encouraging progress in various computer vision tasks. A common belief is that this is attributed to the competence of self-attention in...

The zebra2horse dataset is a widely used benchmark dataset for image-to-image translation tasks in the field of computer vision.

The RIO-10 dataset contains 10 real indoor environments with varying lighting conditions, camera positions, and object arrangements.

The Synthia dataset is a large-scale urban scene understanding dataset, containing 9000 samples. It is used for semantic segmentation tasks.

The dataset used in the paper is ImageNet1K and ImageNet21K for image classification, and MS COCO for object detection.

ImageNet trained PyTorch models are evaluated under various simple image transformations.

Convolutional neural networks (CNNs) have become the state-of-the-art tool for dealing with unsolved problems in computer vision and image processing.

Lung nodule malignancy prediction has been enhanced by advanced deep-learning techniques and effective tricks. Nevertheless, current methods are mainly trained with...

The dataset used in this paper is a video of a house tour, which is used to train a Variational Autoencoder (VAE) to learn a representation map for robot navigation.

Scalability in perception for autonomous driving: Waymo open dataset.

Semantic instance annotation of street scenes by 3D to 2D label transfer.

Text2Pos for city-scale position localization based on textual descriptions. Given a point cloud that represents our surroundings and a query position description, Text2Pos...

The dataset used in the paper for novel-view synthesis from unconstrained image collections.

Places205 is a dataset of 2.5 million images from 205 categories, with 12,000 images per category.

D3Fields — dynamic 3D descriptor fields for zero-shot generalizable robotic manipulation tasks

Single-image shadow removal is a significant task that is still unresolved. Most existing deep learning-based approaches attempt to remove the shadow directly, which can not...

The HO-3D dataset consists of 78,000 images frames on 10 objects.