The ISPRS 2D semantic labeling benchmark (Vaihingen) dataset is used for evaluating the proposed method.

The LoveDA dataset consists of 5,987 high-quality optical remote sensing images with a resolution of 0.3 meters per pixel.

The ISPRS Vaihingen dataset contains 33 high-quality images with topographical information, each with an average resolution of 2494×2064 pixels.

Four widely-used remote sensing datasets are considered for evaluating the efficacy of the proposed approach. Sample images of these datasets are provided in Fig. 5.

The dataset used in the paper is a Delaunay Triangulation with n vertices, where n ranges from 1000 to 10000.

The dataset is a random geometric graph with vertex set [n] based on n i.i.d. random vectors X1,..., Xn drawn from an unknown density f on Rd.

A dataset of pathological OCT scans for photoreceptor layer segmentation

Few-shot segmentation remains challenging due to the limitations of its labeling information for unseen classes. Most previous approaches rely on extracting high-level feature...

The dataset used in the paper is a combination of medical images, including T1, T2, and Flair MRI scans from BraTS, and images from MVTec AD.

A reinterpretation of the U-Net architecture as a modularized structure was proposed as a solution to scale up the segmentation of such images.

The MS COCO dataset is a large benchmark for image captioning, containing 328K images with 5 caption descriptions each.

The dataset used in the paper is not explicitly described. However, it is mentioned that the authors used various benchmarks such as ImageNet, ImageNet ReaL, ImageNet V2, etc.

The dataset used in the paper is a synthetic dataset for off-policy contextual bandits, with contexts x ∈ X, a finite set of actions A, and bounded real rewards r ∈ A → [0, 1].

The Stanford 2D3DS dataset contains omni-directional images.

PASCAL-5i and COCO-20i are datasets used for evaluation of few-shot segmentation.

High-resolution remote sensing images

The dataset used in this paper is a ResNet50 model, which is a convolutional neural network. The model is used to evaluate the effects of quantization on the compute...

The LoveDA dataset contains high-spatial-resolution images from three different cities, focusing on improving the generalization capability of model from different urban and...