Abstract: A transformer-based anomaly detection apparatus includes: an encoder unit including n encoding blocks having a pyramid structure, and extracting respective multi-scale feature maps having the pyramid structure through respective encoding blocks; a spatio-temporal transformer unit having n spatio-temporal transformers, and generating each multi-scale spatio-temporal attention map in which a temporal feature and a spatial feature for the feature map extracted from each encoding block are emphasized through each spatio-temporal transformer; and a decoder unit including n decoders having a reverse pyramid structure, and generating a prediction frame by using outputs of the multi-scale spatio-temporal attention map and a previous layer decoder through n decoders, and an abnormal event is detected by using the prediction frame.

Abstract: A transformer-based anomaly detection apparatus includes: an encoder unit including n encoding blocks having a pyramid structure, and extracting respective multi-scale feature maps having the pyramid structure through respective encoding blocks; a spatio-temporal transformer unit having n spatio-temporal transformers, and generating each multi-scale spatio-temporal attention map in which a temporal feature and a spatial feature for the feature map extracted from each encoding block are emphasized through each spatio-temporal transformer; and a decoder unit including n decoders having a reverse pyramid structure, and generating a prediction frame by using outputs of the multi-scale spatio-temporal attention map and a previous layer decoder through n decoders, and an abnormal event is detected by using the prediction frame.

Abstract: Disclosed are a method and an apparatus for detecting abnormal objects in a video. The method for detecting abnormal objects in a video reconstructs a restored batch by applying each input batch to which an inpainting pattern is applied to a trained auto-encoder model, and fuses a time domain reconstruction error using time domain restored frames output by extracting and restoring a time domain feature point by applying a spatial domain reconstruction error and a plurality of successive frames using a restored frame output by combining the reconstructed restoring batch to a trained LSTM auto-encoder model to estimate an area where an abnormal object is positioned.

Abstract: A device and method for remotely controlling an unmanned aerial vehicle based on reinforcement learning are disclosed. An embodiment provides a device for remotely controlling an unmanned aerial vehicle based on reinforcement learning, where the device includes a processor and a memory connected to the processor, and the memory includes program instructions that can be executed by the processor to determine an inclination direction corresponding to the hand pose of a user, the movement direction of the hand, and the angle in the inclination direction based on sensing data associated with the pose of the hand or the movement of the hand acquired by way of at least one sensor, and determine one of a movement direction, a movement speed, a mode change, a figural trajectory, and a scale of the figural trajectory of the unmanned aerial vehicle according to the determined inclination direction, movement direction, and angle.

Abstract: Disclosed are a method and an apparatus for detecting abnormal objects in a video. The method for detecting abnormal objects in a video reconstructs a restored batch by applying each input batch to which an inpainting pattern is applied to a trained auto-encoder model, and fuses a time domain reconstruction error using time domain restored frames output by extracting and restoring a time domain feature point by applying a spatial domain reconstruction error and a plurality of successive frames using a restored frame output by combining the reconstructed restoring batch to a trained LSTM auto-encoder model to estimate an area where an abnormal object is positioned.

Abstract: A device and method for remotely controlling an unmanned aerial vehicle based on reinforcement learning are disclosed. An embodiment provides a device for remotely controlling an unmanned aerial vehicle based on reinforcement learning, where the device includes a processor and a memory connected to the processor, and the memory includes program instructions that can be executed by the processor to determine an inclination direction corresponding to the hand pose of a user, the movement direction of the hand, and the angle in the inclination direction based on sensing data associated with the pose of the hand or the movement of the hand acquired by way of at least one sensor, and determine one of a movement direction, a movement speed, a mode change, a figural trajectory, and a scale of the figural trajectory of the unmanned aerial vehicle according to the determined inclination direction, movement direction, and angle.

Abstract: A remote control device and method for a UAV, and a motion control device attached to the UAV. The remote control device is carried by a user, allowing the user to remotely control motions of the UAV. A sensor unit generates sensing data by sensing a motion of the remote control device using a sensor. A control unit determines at least one among a direction of inclination of the remote control device, an angle of the direction of inclination, and a period of time for which the direction of inclination is maintained, based on the sensing data, and generating a control command for controlling a motion of the UAV using at least one among the direction of inclination, the angle of the direction of inclination, and the period of time for which the direction of inclination is maintained. A communication unit transmits the control command to the UAV.

Abstract: A remote control device and method for a UAV, and a motion control device attached to the UAV. The remote control device is carried by a user, allowing the user to remotely control motions of the UAV. A sensor unit generates sensing data by sensing a motion of the remote control device using a sensor. A control unit determines at least one among a direction of inclination of the remote control device, an angle of the direction of inclination, and a period of time for which the direction of inclination is maintained, based on the sensing data, and generating a control command for controlling a motion of the UAV using at least one among the direction of inclination, the angle of the direction of inclination, and the period of time for which the direction of inclination is maintained. A communication unit transmits the control command to the UAV.

Abstract: Provided is a growth supplement for efficient production of agricultural and livestock products, more specifically, a growth supplement for environment-friendly crop cultivation and livestock rearing to increase yields of crops and livestock, which is prepared by mixing specific amounts of antibiotic fruit juices, one or more sources of sugar, one or more sources of hydrogen carbonate, one or more sources of chloride ion and antifebrile vegetable juices depending on crops (species) and cultivation (livestock) conditions to feed the livestock or crops. The growth supplement is prepared by admixing and stirring antibiotic fruit juices in an amount of 2 to 10 parts by weight; the source of hydrogen carbonates in an amount of 1 to 4 parts by weight; the source of chloride ion in an amount of 1 to 4 parts by weight; and antifebrile vegetable juices in an amount of 2 to 8 parts by weight, based on 100 parts by the source of sugar.

Abstract: Provided is a growth supplement for efficient production of agricultural and livestock products, more specifically, a growth supplement for environment-friendly crop cultivation and livestock rearing to increase yields of crops and livestock, which is prepared by mixing specific amounts of antibiotic fruit juices, one or more sources of sugar, one or more sources of hydrogen carbonate, one or more sources of chloride ion and antifebrile vegetable juices depending on crops (species) and cultivation (livestock) conditions to feed the livestock or crops. The growth supplement is prepared by admixing and stirring antibiotic fruit juices in an amount of 2 to 10 parts by weight; the source of hydrogen carbonates in an amount of 1 to 4 parts by weight; the source of chloride ion in an amount of 1 to 4 parts by weight; and antifebrile vegetable juices in an amount of 2 to 8 parts by weight, based on 100 parts by the source of sugar.