992 datasets found

The proposed approach is evaluated on three different datasets: MNIST, fashionMNIST, and CIFAR10.

Egocentric video dataset for object detection and segmentation

The dataset used in the paper is the Kaolin-Wisp dataset, which is a benchmark for neural fields research.

The Middlebury 2014 dataset is a benchmark for stereo matching, consisting of 33 pairs of stereo images with sparse depth ground truth.

The VIMER-UFO benchmark dataset consists of 8 computer vision tasks: CPLFW, Market1501, DukeMTMC, MSMT-17, Veri-776, VehicleId, VeriWild, and SOP.

The CAD dataset is a photorealistic 3D generation dataset conditioned on a single image and a text prompt.

MuPoTS-3D is a representative dataset for monocular 3DMPPE, composed of 20 few-second-long sequences with 2–3 people interacting with each other.

The proposed framework for detecting social distancing from video footage

A synthetic fisheye dataset based on the ImageNet-1K, constructed to explore the performance of Transformer models on fisheye images.

Scattering networks are a class of designed Convolutional Neural Networks (CNNs) with fixed weights. We argue they can serve as generic representations for modeling images.

The dataset used for training and testing the proposed deep co-training method for semi-supervised image segmentation.

The dataset used in the paper to evaluate the proposed FG-NeRF method for uncertainty estimation in neural radiance fields.

Deep convolutional neural networks have recently proven extremely competitive in challenging image recognition tasks. This paper proposes the epitomic convolution as a new...

The dataset is used for evaluating the performance of fisheye image rectification algorithms. It contains real-world fisheye images with their corresponding ground truth images...

The dataset is used for fisheye image rectification. It contains synthetic images with fisheye distortion and their corresponding ground truth images, distortion parameters, and...

The dataset used in the paper is a large-scale comparison of pretrained models across computer vision tasks.

The Wrap Wire dataset is a collection of images and labels used for training a visual representation that can bridge the sim2real visual gap.

The Multi-step Pick and Place dataset is a collection of images and labels used for training a visual representation that can bridge the sim2real visual gap.

The Stack Object dataset is a collection of images and labels used for training a visual representation that can bridge the sim2real visual gap.

The Lang4Sim2Real dataset is a collection of image-language pairs used for training a visual representation that can bridge the sim2real visual gap.