Drone footage of endangered dolphins swimming with paddleboarders in Southland could help artificial intelligence, which is being used for conservation. Rodd Trafford filmed the playful pod in Te Waewae Bay, about an hours drive west of Invercargill, on January 9. "I thought it was a once-in-a-lifetime opportunity. I'll never get a chance to film that again," Trafford said. The footage appears to show at least 16 dolphins. READ MORE: * Dolphin advocates say Government's proposed protections are fundamentally flawed * 'It's right in the middle of their hood': Dolphin researcher fearful old dumpsite could spell disaster for Hector's * DOC proposal could cut red tape to building cycle trails on conservation land Trafford said he technically broke drone-use rules when he got the footage, but the Department Of Conservation had since given him the OK because the department would use the footage to enhance its artificial technology work.

The rapid emergence of airborne platforms and imaging sensors are enabling new forms of aerial surveillance due to their unprecedented advantages in scale, mobility, deployment and covert observation capabilities. This paper provides a comprehensive overview of human-centric aerial surveillance tasks from a computer vision and pattern recognition perspective. It aims to provide readers with an in-depth systematic review and technical analysis of the current state of aerial surveillance tasks using drones, UAVs and other airborne platforms. The main object of interest is humans, where single or multiple subjects are to be detected, identified, tracked, re-identified and have their behavior analyzed. More specifically, for each of these four tasks, we first discuss unique challenges in performing these tasks in an aerial setting compared to a ground-based setting. We then review and analyze the aerial datasets publicly available for each task, and delve deep into the approaches in the aerial literature and investigate how they presently address the aerial challenges. We conclude the paper with discussion on the missing gaps and open research questions to inform future research avenues.

An air attack in Ethiopia's northern region of Tigray on Monday killed at least 17 people, mostly women, and wounded dozens in the town of Mai Tsebri, two aid workers have told Reuters news agency, citing local authorities and witnesses. Monday's attack came on the day US President Joe Biden in a phone call raised concerns with Ethiopian Prime Minister Abiy Ahmed about civilian casualties and suffering caused by air attacks. At least 56 people were killed and 30 injured, including some children, in a drone attack on a camp for displaced people in Tigray on Friday. A report by the zonal administration said women at a flour mill made up most of the casualties in Monday's drone attack, a source who saw the report told The Associated Press news agency. The source spoke on condition of anonymity because they were not authorised to speak about it to reporters.

Taiwan's parliament has passed an extra spending bill of $8.6bn, in the latest bid to boost defence capabilities against growing military threats from China. Lawmakers on Tuesday agreed unanimously to pass the special budget, which comes on top of a record annual defence budget of about $17bn set for 2022. The additional resources are aimed at acquiring precision missiles and mass-manufacture high-efficiency naval ships "in the shortest period of time" to boost the island's sea and air capabilities, the government said. The budget includes a coastal anti-ship missiles system, a locally developed Wan Chien (Ten Thousand Swords) cruise missile as well as an attack drone system and installation of combat systems on coastguard ships. Democratic Taiwan lives under constant threat of an invasion by authoritarian China, which claims the self-ruled island as part of its territory.

Kyiv, Ukraine – With their country's fate being discussed at this week's US-Russia talks, Ukrainians took to the streets over the weekend to defend their independence and champion an additional cause – that of Kazakhstan's protests. On Sunday, demonstrators in Kyiv and Kharkov, Ukraine's second largest city, held signs that read "Say no to Putin" and flew Kazakhstan flags alongside Ukrainian ones. Kazakhstan's blue and gold flag also appeared in the winter skies over Kyiv on Saturday, flown from a drone in an act of protest organised by Dronarium, an unmanned aerial vehicle enthusiast community known for political statements. "Every nation has the right to protect their socioeconomic and political rights through peaceful protest," said drone operator Vitaly Shevchuk."We After a week of violent protests that began over a hike in fuel prices and quickly spread across the country – leaving at least 164 dead, 2,000 injured and almost 6,000 arrested – a Russian-led military alliance has now restored control of Kazakhstan to the government. The Collective Security Treaty Organization (CSTO), an alliance of several former Soviet states, deployed about 2,500 troops to Kazakhstan to help quell the protest, including Russian paratroopers who are guarding "vital facilities and social infrastructure," a Russian Defence Ministry statement said. Critics have accused Russia of "occupation" over its involvement, with Kazakh Mukhtar Ablyazov, a former minister turned opposition leader, warning President Vladimir Putin will draw the country into "a structure like the Soviet Union" unless the West intervenes. Motivated more by their own hopes to defy Putin than in sharing a common cause with the protests, Ukrainians also urged resistance. "The dictator [Putin] wants to rebuild the USSR by force," said Olga Angelova, who was among the protesters in Kyiv. "He must be stopped – we Ukrainians will resist the occupiers.

Supermaneuverability, in broad terms, refers to the complex forms of non-conventional maneuverability that are found in high-performance combat aircraft. This capability includes maneuvers such as the Pugachev cobra, Kulbit and Herbst maneuver [1-3]; as well as broader, competing, classifications of flight behavior, including rapid nose-pointing-and-shooting (RaNPAS), pure sideslip maneuvering (PSM) [4,5] and direct force Page 3 of 32 maneuvering (DFM) [6]. The development of supermaneuverable aircraft has been founded on advances in the study of unstable airframes, and the development of vectored propulsion technology [1,2]. Modern supermaneuverable aircraft remain characterized by these mechanisms; but increasing interdisciplinary contact with biological studies of maneuverability in flying creatures has led to parallel studies of an alternative, biomimetic, mechanism of supermaneuverability: one based on controlled wing morphing and motion. Thus far, biomimetic perching in unmanned aerial vehicles (UAVs) has been a central focus of these studies [7-9], with extensions into hover-to-cruise transition maneuvers [10], and incidence-based stall turns [11]. These maneuvers are primarily bio-inspired, and as such, studies of the biomimetic mechanism supermaneuverability have remained disjointed from studies of the thrust-vectored mechanism: the relationships between biomimetic and thrustvectored maneuvers, mechanisms, and capabilities are rarely recognized [3].

One of the important challenges of autonomous flight is the Sense and Avoid (SAA) task to maintain enough separation from obstacles. While the route of an autonomous drone might be carefully planned ahead of its mission, and the airspace is relatively sparse, there is still a chance that the drone will encounter unforeseen airborne objects or static obstacles during its autonomous flight. The autonomous SAA module has to take on the tasks of situational awareness, decision making, and flying the aircraft, while performing an evasive maneuver. There are several alternatives for onboard sensing including radar, LIDAR, passive electro-optical sensors, and passive acoustic sensors. Solving the SAA task with visual cameras is attractive because cameras have relatively low weight and low cost. For the purpose of this challenge, we consider a solution that solely relies on a single visual camera and Computer Vision technique that analyzes a monocular video. Flying airborne objects pose unique challenges compared to static obstacles. In addition to the typical small size, it is not sufficient to merely detect and localize those objects in the scene, because prediction of the future motion is essential to correctly estimate if the encounter requires a collision avoidance maneuver and create a safer route. Such prediction will typically rely on analysis of the motion over a period of time, and therefore requires association of the detected objects across the video frames.

Imagine flying a small, robotic aircraft from goal post to goal post on an American football field. Now, repeat the flight 470 more times, and you'll match the record-setting 32-mile autonomous drone flight recorded by Aakash Sinha's industry-leading startup based in New Delhi. "It's only the beginning," said Sinha, a 2003 School of Computer Science graduate with a master's degree in robotics. "I'm super excited about how drones can change things, not just here in India but globally." From delivering vaccines in hard-to-reach areas to limiting fossil fuel leaks in expansive pipelines, the possibilities for positive change are endless.

Japan overturned in 2020 its decision to cancel acquisition of U.S.-made reconnaissance drones due to their massive costs out of consideration to then-U.S. President Donald Trump, who was promoting American weapons exports, according to sources close to the matter. The government of then-Prime Minister Shinzo Abe had told Washington in the spring of 2020 that it would not purchase the Global Hawk drones, but reversed the decision in the summer after Tokyo scrapped in June that year its planned deployment of U.S.-developed land-based Aegis Ashore ballistic missile defense systems, they said. The about-face was prompted by concerns that cancellation of the Global Hawk acquisition would "anger Mr. Trump, who has insisted on exporting U.S.-made weapons," according to a source familiar with the matter. The policy change reflected "excessive consideration for Mr. Trump," the source said.

Compared with existing vehicle re-identification (ReID) tasks conducted with datasets collected by fixed surveillance cameras, vehicle ReID for unmanned aerial vehicle (UAV) is still under-explored and could be more challenging. Vehicles with the same color and type show extremely similar appearance from the UAV's perspective so that mining fine-grained characteristics becomes necessary. Recent works tend to extract distinguishing information by regional features and component features. The former requires input images to be aligned and the latter entails detailed annotations, both of which are difficult to meet in UAV application. In order to extract efficient fine-grained features and avoid tedious annotating work, this letter develops an unsupervised self-aligned network consisting of three branches. The network introduced a self-alignment module to convert the input images with variable orientations to a uniform orientation, which is implemented under the constraint of triple loss function designed with spatial features. On this basis, spatial features, obtained by vertical and horizontal segmentation methods, and global features are integrated to improve the representation ability in embedded space. Extensive experiments are conducted on UAV-VeID dataset, and our method achieves the best performance compared with recent ReID works.