Computer Vision
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:
1143 Views
The dataset is generated specifically to simulate the essential components for driving environments in a virtual campus of Chulalongkorn University, including street blocks, various pavements, lane markings, traffic signs, lamp poles, and pedestrians, among other features. We selected this campus for our simulation due to its distinctive road and pavement environments, which are unique to Thailand and other Asian countries. This choice contrasts with many widely cited datasets that predominantly feature environments from European or other regions.
- Categories:
129 Views
Falling objects from buildings can cause severe injuries to pedestrians due to the great impact force they exert. Although surveillance cameras are installed around some buildings, it is challenging for humans to capture such events in surveillance videos due to the small size and fast motion of falling objects, as well as the complex background. Therefore, it is necessary to develop methods to automatically detect falling objects around buildings in surveillance videos.
- Categories:
227 Views
While deep learning has catalyzed breakthroughs across numerous domains, its broader adoption in clinical settings is inhibited by the costly and time-intensive nature of data acquisition and annotation. To further facilitate medical machine learning, we present an ultrasound dataset of 10,223 Brightness-mode (B-mode) images consisting of sagittal slices of porcine spinal cords (N=25) before and after a contusion injury.
- Categories:
24 Views
The removal of surgical tools from the brain is a critical aspect of post-operative care. Surgical sponges such as cotton balls are one of the most commonly retained tools, as they become visually indistinguishable from the surrounding brain tissue when soaked with blood and can fragment into smaller pieces. This can lead to life-threatening immunological responses and invasive reoperation, demonstrating the need for new foreign body object detection methods.
- Categories:
46 Views
Animal habitat surveys play a critical role in preserving the biodiversity of the land. One of the effective ways to gain insights into animal habitats involves identifying animal footprints, which offers valuable information about species distribution, abundance, and behavior.
- Categories:
220 Views
In order to train the joint contrastive representation learning module, we constructe a large Text Annotated Distortion, Appearance and Content (TADAC) image database.
- Categories:
380 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:
641 Views
The dataset consists of around 335K real images equally distributed among 7 classes. The classes represent different levels of rain intensity, namely "Clear", "Slanting Heavy Rain", "Vertical Heavy Rain", "Slanting Medium Rain", "Vertical Medium Rain", "Slanting Low Rain", and "Vertical Low Rain". The dataset has been acquired during laboratory experiments and simulates a low-altitude flight. The system consists of a visual odometry system comprising a processing unit and a depth camera, namely an Intel Real Sense D435i.
- Categories:
236 Views
Most existing superpixel algorithms only consider color intensity and position coordinates, while ignoring local neighborhood factors. This limitation leads to low applicability in noisy and cluttered environments. To address this issue, we propose a seminal and novel Fuzzy C-Means clustering with Region Constraints for Superpixel generation (RCFCMS). First, employing region constraints to prevent boundary crossing. Second, adopting spatial information to mitigate noise interference. Third, utilizing soft membership to convert labels.
- Categories:
16 Views