GoPro Hero Research Articles

Calibration of action cameras for photogrammetric purposes.

The use of action cameras for photogrammetry purposes is not widespread due to the fact that until recently the images provided by the sensors, using either still or video capture mode, were not big enough to perform and provide the appropriate analysis with the necessary photogrammetric accuracy. However, several manufacturers have recently produced and released new lightweight devices which are: (a) easy to handle, (b) capable of performing under extreme conditions and more importantly (c) able to provide both still images and video sequences of high resolution. In order to be able to use the sensor of action cameras we must apply a careful and reliable self-calibration prior to the use of any photogrammetric procedure, a relatively difficult scenario because of the short focal length of the camera and its wide angle lens that is used to obtain the maximum possible resolution of images. Special software, using functions of the OpenCV library, has been created to perform both the calibration and the production of undistorted scenes for each one of the still and video image capturing mode of a novel action camera, the GoPro Hero 3 camera that can provide still images up to 12 Mp and video up 8 Mp resolution.

A comparison of multi-view 3D reconstruction of a rock wall using several cameras and a laser scanner

Abstract. This work presents a comparative study between multi-view 3D reconstruction using various digital cameras and a terrestrial laser scanner (TLS). Five different digital cameras were used in order to estimate the limits related to the camera type and to establish the minimum camera requirements to obtain comparable results to the ones of the TLS. The cameras used for this study range from commercial grade to professional grade and included a GoPro Hero 1080 (5 Mp), iPhone 4S (8 Mp), Panasonic Lumix LX5 (9.5 Mp), Panasonic Lumix ZS20 (14.1 Mp) and Canon EOS 7D (18 Mp). The TLS used for this work was a FARO Focus 3D laser scanner with a range accuracy of ±2 mm. The study area is a small rock wall of about 6 m height and 20 m length. The wall is partly smooth with some evident geological features, such as non-persistent joints and sharp edges. Eight control points were placed on the wall and their coordinates were measured by using a total station. These coordinates were then used to georeference all models. A similar number of images was acquired from a distance of between approximately 5 to 10 m, depending on field of view of each camera. The commercial software package PhotoScan was used to process the images, georeference and scale the models, and to generate the dense point clouds. Finally, the open-source package CloudCompare was used to assess the accuracy of the multi-view results. Each point cloud obtained from a specific camera was compared to the point cloud obtained with the TLS. The latter is taken as ground truth. The result is a coloured point cloud for each camera showing the deviation in relation to the TLS data. The main goal of this study is to quantify the quality of the multi-view 3D reconstruction results obtained with various cameras as objectively as possible and to evaluate its applicability to geotechnical problems.

A fresh look at hypomagnesaemia

Humans use food rewards as positive reinforcement for training horses, but there is little evidence to show that human interaction (scratching or patting) has reward value or if domestic horses perceive human touch as social bonding. Most equine training is based on negative reinforcement, but food is a known reward for training through positive reinforcement. This study looked at how horses perceive human interaction as a form of positive reward by examining whether scratching and patting can serve as a reward for a behavior and how this compares to known rewards (treats) based on horses’ ability to use symbols to show preferences. A secondary purpose of this study was to explore behaviors before, during, and after the choice test and code changes based on known equine ethology and determine any correlations between known individual histories of horses, their behavior, and any differences in preferences for human interactions based on whether the human was familiar versus unfamiliar. For the quantitative component, the study counted the number of times each horse touched each target (touch counts) and behaviors were recorded with use of a GoPro Hero camera. The touch counts for each symbol for the last trial were compared using Observation Oriented Modelling (OOM), a non-parametric approach to analyze patterns in data, and the behaviors were coded and compared to known behaviors in equine ethology. Post-hoc analysis of final touch counts resulted in a pattern of fewer than 2 touches for both scratches and pats for most horses and consistently ten touches (the maximum) for treats for all horses. Post-hoc pattern analysis resulted in a Percent Correct Classification (PCC) index of 93.94 (c-value <.001) for the familiar human and 87.88 (c-value <.001) for the unfamiliar researcher suggesting that, given the study conditions, all horses in this study showed preference for treats over human contact, regardless of handling or training histories. All horses also displayed arousal behaviors during the study suggesting the study, and potentially the novel food component, induced an arousal response.