63 datasets found

The dataset used for image recognition using deep convolutional neural networks.

The dataset used for image recognition using deep convolutional neural networks.

Conformer is a dual network structure that combines CNN-based local features with transformer-based global representations for enhanced representation learning.

The PASCAL VOC dataset contains 20 classes, including person, animal, vehicle, and indoor, with 9,963 images containing 24,640 annotated objects.

ADE20K: A Dataset for Semantic Segmentation

MS COCO: Common Objects in Context

MMDetection: Open MMLab Detection Toolbox and Benchmark

ConvMLP: a Hierarchical Convolutional MLP backbone for visual recognition

The MIT Indoor Scene Recognition dataset contains 67 categories of indoor scenes.

The dataset used in this paper is the VGG Network E, a deep convolutional neural network for image recognition.

The dataset consists of binary images of sprites with variations in the shape (oval, square, and heart) and four geometric factors: scale (6 variation modes), rotation (40), and...

Deep Image: Scaling up image recognition

A dataset of images of holy places (Kaaba, Zamzam, Maqam Ibrahim) for training a deep learning model.

The street view house number recognition task involves transcribing an image with house numbers to a string of digits.

RoadTracer is an automatic recognition method of road network system, called RoadTracer. RoadTracer can generate a road map on the ground surface from aerial photograph data.

The dataset used in the paper is Places, a large dataset of 400k pairs of images from the Places 205 dataset and corresponding spoken audio captions.

Training neural networks on image datasets generally require extensive experimentation to find the optimal learning rate regime.

The dataset used in the paper is the Imagenet dataset, which is a large annotated dataset of images used for image classification tasks.

The paper proposes a Towards Fair Knowledge Transfer (TFKT) framework to handle the fairness challenge in imbalanced cross-domain learning.

Binary images are simple — two possible pixel-valued signals with a single channel. The simplicity of binary images has a significant advantage compared to colored and gray...