992 datasets found

DP-NeRF is a novel NeRF framework from blurry inputs, that imposes two physical priors to effectively construct a clean NeRF.

Brain lesion segmentation provides a valuable tool for clinical diagnosis, and convolutional neural networks (CNNs) have achieved unprecedented success in the task. Data...

A new dataset that contains demanding illumination conditions and provides a rich set of sensory data for autonomous driving.

RevColV2 is a new architecture that learns disentangled representations during MIM pre-training and keeps a unified autoencoder architecture when transferring to downstream tasks.

ACCO: Automated Causal CNN Scheduling Optimizer for Real-Time Edge Accelerators

The dataset used in the paper is the LLFF dataset, which contains real-world scenes and is used for training and testing the proposed neural radiance field model.

The LiTS17 and SLiver07 datasets are used for training and testing the proposed SAR-U-Net method.

Monocular depth estimation dataset

A large-scale benchmark for evaluating the brain-likeness of neural networks.

A shallow recurrent model that follows neuroanatomy more closely than standard machine learning ANNs, and is among the top models on Brain-Score yet remains competitive on...

Nuscenes is a large-scale autonomous driving dataset that provides high-resolution BEV semantic occupancy labels for roads and vehicles.

Depth estimation, visual odometry, and bird’s-eye-view scene layout estimation present three critical tasks for driving scene perception, which is fundamental for motion...

The dataset used in the paper is a text-to-image synthesis dataset.

The dataset used in the paper for cinematic motion transfer, consisting of NeRF representations of 3D scenes and reference video clips.

This dataset is used to validate the effectiveness of the SCINeRF method for recovering 3D scenes from a single snapshot compressed image.

This dataset is used to validate the effectiveness of the SCINeRF method for recovering 3D scenes from a single snapshot compressed image.

CLIP-PAE: Projection-Augmentation Embedding to Extract Relevant Features for a Disentangled, Interpretable and Controllable Text-Guided Face Manipulation

The dataset used in this paper is a collection of images, and the authors used it to train and evaluate Variational Autoencoders (VAEs).

This work investigates the robustness of SAM to corruptions and adversarial attacks.

SAM is a foundation model in computer vision that can segment anything.