Information on trees at the individual level is crucial for monitoring forest ecosystems and planning forest management. Current monitoring methods involve ground measurements, requiring extensive cost, time and labour.

Detecting small targets in drone imagery is challenging due to low resolution, complex backgrounds, and dynamic scenes. We propose EDNet, a novel edge-target detection framework built on an enhanced YOLOv10 architecture, optimized for real-time applications without post-processing. EDNet incorporates an XSmall detection head and a Cross Concat strategy to improve feature fusion and multi-scale context awareness for detecting tiny targets in diverse environments. Our unique C2f-FCA block employs Faster Context Attention to enhance feature extraction while reducing computational complexity. The WIoU loss function is employed for improved bounding box regression. With seven model sizes ranging from Tiny to XL, EDNet accommodates various deployment environments, enabling local real-time inference and ensuring data privacy. Notably, EDNet achieves up to a 5.6% gain in mAP@50 with significantly fewer parameters. On an iPhone 12, EDNet variants operate at speeds ranging from 16 to 55 FPS, providing a scalable and efficient solution for edge-based object detection in challenging drone imagery. The source code and pre-trained models are available at: https://github.com/zsniko/EDNet.

Animal populations worldwide are rapidly declining, and a technology that can accurately count endangered species could be vital for monitoring population changes over several years. This research focused on fine-tuning object detection models for drone images to create accurate counts of animal species. Hundreds of images taken using a drone and large, openly available drone-image datasets were used to fine-tune machine learning models with the baseline YOLOv8 architecture. We trained 30 different models, with the largest having 43.7 million parameters and 365 layers, and used hyperparameter tuning and data augmentation techniques to improve accuracy. While the state-of-the-art YOLOv8 baseline had only 0.7% accuracy on a dataset of safari animals, our models had 95% accuracy on the same dataset. Finally, we deployed the models on the Jetson Orin Nano for demonstration of low-power real-time species detection for easy inference on drones.

Nowadays, there is a wide availability of datasets that enable the training of common object detectors or human detectors. These come in the form of labelled real-world images and require either a significant amount of human effort, with a high probability of errors such as missing labels, or very constrained scenarios, e.g. VICON systems. On the other hand, uncommon scenarios, like aerial views, animals, like wild zebras, or difficult-to-obtain information, such as human shapes, are hardly available. To overcome this, synthetic data generation with realistic rendering technologies has recently gained traction and advanced research areas such as target tracking and human pose estimation. However, subjects such as wild animals are still usually not well represented in such datasets. In this work, we first show that a pre-trained YOLO detector can not identify zebras in real images recorded from aerial viewpoints. To solve this, we present an approach for training an animal detector using only synthetic data. We start by generating a novel synthetic zebra dataset using GRADE, a state-of-the-art framework for data generation. The dataset includes RGB, depth, skeletal joint locations, pose, shape and instance segmentations for each subject. We use this to train a YOLO detector from scratch. Through extensive evaluations of our model with real-world data from i) limited datasets available on the internet and ii) a new one collected and manually labelled by us, we show that we can detect zebras by using only synthetic data during training. The code, results, trained models, and both the generated and training data are provided as open-source at https://eliabntt.github.io/grade-rr.

We enjoyed messing around with the games, but pretty quickly our mind wandered to what these things can do. VR is still a young space, and you quite frequently find yourself wanting an app that doesn't exist. At the top of our list: Could you view the world through a drone's camera while you flew it? We'd taken a few AI courses in the past, and we thought that single camera 3D VR might just be possible. In the world of autonomous vehicles, there is similar work underway in the form of research into "Pseudo-LIDAR".

Since the death of George Floyd on May 25, Americans have taken to the streets to peacefully protest in unprecedented numbers, calling for an end to our national culture of racism and police brutality. These protests have, on too many occasions, been met with violent force from police, who have been caught on camera using tear gas, pepper spray, rubber bullets, and other supposedly less-lethal weapons against unarmed and compliant people. Police around the country are also devoting considerable time and energy to collecting intelligence on protesters and protest movements, with methods ranging from monitoring social media posts to aerial surveillance--sometimes, with drones. Police, military, and federal government forces have regularly flown surveillance helicopters and small, crewed surveillance aircraft over protest areas, capturing real-time video and photographs of protest movements. The New York Times found that by mid-June, the Department of Homeland Security had captured more than 270 hours of surveillance footage of protests from helicopters, airplanes, and drones, data that was shared with a digital network accessible by other federal agencies and by police departments.

In areas like the Caribbean that face considerable risk from natural hazards like earthquakes, hurricanes, and floods, these forces of nature can have a devastating effect. This is especially true where houses and buildings are not up to modern construction standards, often in poor and informal settlements. While buildings can be retrofit to better prepare them for disaster, the traditional method for identifying high-risk buildings involves going door to door by foot, taking weeks if not months and costing millions of dollars. This is where AI can help. WeRobotics and the World Bank Global Program for Resilient Housing have teamed up to prepare aerial drone imagery of buildings across the Caribbean annotated with characteristics that matter to building inspectors.

For years, humans have recognized images better than computers. Our error rate has been steadily at 5 percent while computer algorithms were at 30 percent. However, with the rise of computer vision and deep learning, the gap between humans and computers has slowly closed. Within the last two years, researchers have seen computer algorithms show an error rate of less than 5 percent, surpassing humans. These advancements bring significant potential to many different industries. In the infrastructure industry, users have applied reality modeling in countless projects to improve all workflows.

On Monday, President Trump signed the the $717 billion annual National Defense Authorization Act, which was easily passed by Congress in weeks prior. Much attention has understandably been placed on big-ticket items like $7.6 billion for acquiring 77 F-35 fighters, $21.9 billion for the nuclear weapons program, and $1.56 billion for three littoral combat ships--despite the fact that the Navy requested only one in the budget. What has gotten less attention is how the bill cements artificial intelligence programs in the Defense Department and lays the groundwork for a new national-level policy and strategy in the form of an artificial intelligence commission. As artificial intelligence and machine learning algorithms are integrated into defense technology, spending on these technologies is only going to increase in years to come. While spending for many AI programs in the NDAA is in the tens of millions at present, one budget for a project that did not go through the normal appropriations process could have a total cost of $1.75 billion over the next seven years.

Man has always been fascinated with a view of the world from the top -- building watch-towers, high fortwalls, capturing the highest mountain peak. To capture a glimpse and share it with the world, people went to great lengths to defy gravity, enlisting the help of ladders, tall buildings, kites, balloons, planes, and rockets. Today, access to drones that can fly as high as 2kms is possible even for the general public. These drones have high resolution cameras attached to them that are capable of acquiring quality images which can be used for various kinds of analysis. With easier access to drones, we're seeing a lot of interest and activity by photographers & hobbyists, who are using it to make creative projects such as capturing inequality in South Africa or breathtaking views of New York which might make Woody Allen proud.