datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
agri-vision-2021-segmentation006_behavior_dataset_0_8999vada-sambhar
Dataset Card for vada-sambhar
Dataset Description
The dataset contains of images of my favorite south indian dish - Vada Sambhar.
Dataset Curators
The data has been downloaded from Google images.
Licensing Information
The vada-sambhar dataset version 1.0.0 is released under the Apache-2.0 License.
VADE-DatasetThis repository is associated with the paper VADE: Variance-Aware Dynamic Sampling via Online Sample-Level Difficulty Estimation for Multimodal RL.
VADE (Variance-Aware Dynamic Sampling via Online Sample-Level Difficulty Estimation) is a framework designed to enhance multimodal reinforcement learning. It addresses the critical gradient vanishing problem in group-based policy optimization by dynamically selecting informative samples, thereby amplifying training signals while eliminating extra… See the full description on the dataset page: https://huggingface.co/datasets/FloSophorae/VADE-Dataset.wrap_1Cdreambooth_hackathon_vada_pavgrasp_multiObject_upgrade
Multi-Object Grasp Detection Dataset
Dataset for robotic grasping with multiple objects per scene. Contains RGB images, depth maps, and grasp annotations for various objects.
Dataset Details
Total samples: 96 scenes
Objects: Various household items (screwdriver, pliers, scissors, wrench, etc. - in russian)
Grasps per scene: Multiple grasp poses per object
Image resolution: 640x480 pixels
Grasp Annotation Format. Each grasp is represented as a quadrilateral with 4… See the full description on the dataset page: https://huggingface.co/datasets/vadim-vlm4vla/grasp_multiObject_upgrade.ae5_vadim_esh_alch_vision_datasetVADdatasetAgriculture_Visionvadimshatilovchromav1-dataset