*.jpg (zip) *.png (zip)*.txt (zip)
In recent years, the success of large-scale visionlanguage models (VLMs) such as CLIP has led to their increased usage in various computer vision tasks. These models enable zero-shot inference through carefully crafted instructional text prompts without task-specific supervision. However, the potential of VLMs for generalization tasks in remote sensing (RS) has not been fully realized. To address this research gap, we propose a novel image-conditioned prompt learning strategy called the Visual Attention Parameterized Prompts Learning Network (APPLeNet).
- Categories:
64 Views
These are tight pedestrian masks for the thermal images present in the KAIST Multispectral pedestrian dataset, available at https://soonminhwang.github.io/rgbt-ped-detection/
Both the thermal images themselves as well as the original annotations are a part of the parent dataset. Using the annotation files provided by the authors, we develop the binary segmentation masks for the pedestrians, using the Segment Anything Model from Meta.
- Categories:
457 Views
To address the international community’s needs for the safeguarding and monitoring of enriched uranium materials and to prevent the illicit theft, destruction, and transfer of such materials, it is crucial to explore non-destructive analytical techniques for measuring uranium content in enriched uranium materials. Currently, the Am-Li source is commonly used as an excitation source in active uranium measurement methods.
- Categories:
188 Views
CT (roentgen-ray computed tomography) A beam of x-rays is shot straight through the brain. As it comes out the other side, the beam is blunted slightly because it has hit dense living tissues on the way through. Blunting or "attenuation" of the x-ray comes from the density of the tissue encountered along the way. Very dense tissue like bone blocks lots of x-rays; grey matter blocks some and fluid even less. X-ray detectors positioned around the circumference of the scanner collect attenuation readings from multiple angles. A computerized algorithm reconstructs an image of each slice
- Categories:
399 Views
The advancement of machine and deep learning methods in traffic sign detection is critical for improving road safety and developing intelligent transportation systems. However, the scarcity of a comprehensive and publicly available dataset on Indian traffic has been a significant challenge for researchers in this field. To reduce this gap, we introduced the Indian Road Traffic Sign Detection dataset (IRTSD-Datasetv1), which captures real-world images across diverse conditions.
- Categories:
1059 Views
This dataset is a valuable resource for those working on daylight illumination-related projects. It includes a comprehensive collection of images captured under various daylight conditions. These images can be used for tasks such as illumination estimation, scene relighting, and object insertion. The dataset features scenes illuminated by different intensities and angles of daylight, providing a rich set of examples for realistic daylight simulation.
- Categories:
176 Views
The LuFI-RiverSnap dataset includes close-range river scene images obtained from various devices, such as UAVs, surveillance cameras, smartphones, and handheld cameras, with sizes up to 4624 × 3468 pixels. Several social media images, which are typically volunteered geographic information (VGI), have also been incorporated into the dataset to create more diverse river landscapes from various locations and sources.
Please see the following links:
- Categories:
628 Views
This dataset contains MRI scans from 5 MS and 5 NMO cases from the Universiti Teknologi MARA (UiTM) hospital Malaysia. The brain lesions in the MRI scans have been annotated by a consultant radiologist from Pakistan Institute of Medical Sciences (PIMS) Islamabad Pakistan. The ground truth lesion masks are available as png files, whereas the brain scans are available as jpg files.
- Categories:
385 Views
We conducted a retrospective collection, covering 167 children who were examined and treated at the Children's Hospital of Chongqing Medical University from March 12, 2014 to January 7, 2022, with a total of 1634 IRI image sequences. This study has been registered with the Chinese Clinical Trial Registry, registration number ChiCTR2200058971, and complied with the provisions of the Declaration of Helsinki (DoH). The study was approved by the Institutional Ethical Review Board (document number 2022,69), and a waiver of informed consent was obtained.
- Categories:
199 Views
The Deepfake-Synthetic-20K dataset significantly contributes to digital forensics and deepfake detection research. It comprises 20,000 high-resolution, synthetic human face images generated using the advanced StyleGAN-2 architecture. This dataset is designed to support the development and evaluation of machine-learning models that can differentiate between real and artificially synthesized human faces. Each image in the dataset has been meticulously crafted to ensure a diverse representation of age, gender, and ethnicity, reflecting the variability seen in global human populations.
- Categories:
1376 Views