Abstract: A training data generation apparatus may include a first network and a second network that individually learn a first image based on supervised learning and perform ensemble learning on a second image based on unsupervised learning. The apparatus also may include a fusion network that obtains a fusion output value based on the ensemble learning results of the first network and the second network. The apparatus also may include a target network that learns the second image to imitate the fusion output value.

Abstract: An apparatus for building a deep learning model for image learning includes a first deep learning model configured to obtain a first spatial feature map and a first heatmap including center information of an object belonging to an image by learning the image. The apparatus also includes a second deep learning model configured to perform imitation learning on the first deep learning model. The second deep learning model may obtain a second spatial feature map by learning the image and perform learning such that the second spatial feature map imitates the first spatial feature map. The second deep learning model may also obtain a second heatmap including a center of the object and perform learning such that the second heatmap imitates the first heatmap.

Abstract: The present disclosure relates to a vehicle rear seat including: a center seat; and side seats located on the left and right of the center seat, wherein, the center seat is capable of protruding by moving the center seat forward with respect to the side seats, and in the state in which the center seat protrudes forward, it is possible to increase an inter-passenger distance so that physical contact between the passenger of the center seat and the passenger of each of the side seats can be prevented as much as possible.

Abstract: A seat for a vehicle with a cushion tip-up and down function is provided. The seat is configured such that a cushion leg provided in a cushion is held, in a fitted manner, by a support spring of a support structure provided in a seat base, and when the cushion is rotated, the cushion leg is allowed to be folded and ejected by being automatically rotated, thereby improving the convenience of use.

Abstract: A vehicle seat includes a fixed frame installed in a vehicle body, a rotating link connected to the fixed frame through a rotation shaft and rotating relative to the rotation shaft, a seat cushion frame having one side of a rear end portion hinge-connected to the rotating link at a first connection point to rotate the rotating link when rotating, and a seat back frame having one side of a lower end portion hinge-connected to the fixed frame at a second connection point and the other side of the lower end portion hinge-connected to the other side of the rear end portion of the seat cushion frame at a third connection point to rotate relative to the fixed frame when the seat cushion frame is rotated.

Abstract: An apparatus for generating training data includes a communication device that receives a daytime road image, a memory storing training data generation model for generating a night composite image from the daytime road image, and a processor that degrades strength of the daytime road image based on that the daytime road image is input as an input value to the training data generation model and reflects a night feature to generate the night composite image.

Abstract: A seat for a vehicle is proposed. The seat for a vehicle is configured to perform sliding of a seat cushion in conjunction with reclining of a seatback without a separate driver part or recliner, and to realize tipping-up of the seat cushion with the seat cushion and the seatback removably provided, and to be easily mounted and assembled to a vehicle to secure the inside space utilization of the vehicle and achieve cost reduction.

Abstract: A learning device for generating a composite image similar to a real image, a learning method, a testing device for generating a training dataset, and a testing method are provided. The learning device includes a composite image generation network that applies a convolution operation to a simulation image at least once to generate a composite image. The learning device has a first discrimination network that applies a first discrimination operation to a first input image to generate a first discrimination score and has a second discrimination network that applies a second discrimination operation to a second input image to generate a second discrimination score. The learning device also includes a parameter update device that performs back-propagation of a first loss generated with reference to at least a portion of the first discrimination score or the second discrimination score to update a parameter of the composite image generation network.

Abstract: An object region segmentation device and an object region segmentation method thereof are provided. The object region segmentation device includes a processor and storage. The storage stores a deep-learning network model for segmenting an object region in an image. The deep-learning network model includes a first network model for generating a pseudo label, a second network model for generating a confidence map for the pseudo label, and a third network model for segmenting the object region in the image. The processor inputs an unlabeled image to the first network model to generate the pseudo label, inputs the pseudo label to the second network model to generate the confidence map, and trains the third network model using a pseudo label corresponding to at least one pixel, a confidence level of which is greater than or equal to a threshold, on the confidence map.

Abstract: A vehicle seat includes first and second sub-seatbacks also provided as first and second seat cushions mounted on a main seatback to be rotatable in forward-rearward directions. A forward facing position in which an occupant is accommodated while viewing forward and a rearward facing position in which the occupant is accommodated while viewing rearward is easily realized, in response to forward and rearward rotation of the first sub-seatback and the second sub-seatback also provided as first and second seat cushions. Furthermore, a full-bed mode in which the seat is horizontally spread like a bed, a cushion tip-up mode in which the entirety of the seat is positioned upright for ensuring freight loading and an occupant entering and exiting passage, and the like are realized.

Abstract: An apparatus for extracting image noise includes an encoder that encodes a first image and a second image in a time-series image to output an encoding feature map. The apparatus also includes a noise information extraction layer that generates a first noise prototype vector to an M-th noise prototype vector of the first image and the second image, and outputs first noise information with reference to at least a part of the first noise prototype vector to the M-th noise prototype vector and the encoding feature map. The apparatus additionally includes a decoder that decodes the first noise information to output second noise information. The apparatus further includes a parameter updater that outputs a loss with reference to the first image, the second image, and the corresponding second noise information, and updates parameters of the encoder, the noise information extraction layer and/or the decoder through back propagation using the loss.

Abstract: A multi-task learning device includes a feature extraction layer that generates a first feature corresponding to a first image and a second feature corresponding to a second image; a first decoding layer that generates a first task inference result corresponding to the first image; a second decoding layer that generates a second task inference result corresponding to the second image; a first loss layer that generates a first task loss with reference to the first task inference result and a first task ground truth (GT) result corresponding to the first task inference result; a second loss layer that generates a second task loss with reference to the second task inference result and a second task GT result corresponding to the second task inference result; a feature loss layer that generates a feature loss with reference to the first feature and the second feature; and a parameter updater that updates parameters of at least some of the various layers.

Abstract: Machine learning and testing devices are provided. The device applies a neural network operation of a first student neural network to a training image to generate first prediction information, applies a neural network operation of a second student neural network to generate second prediction information, applies an error identification operation to a first integrated image to generate first error identification prediction information, applies the error identification operation to a second integrated image to generate second error identification prediction information, applies a network operation of a first teacher neural network to the training image to generate first pseudo label information, applies a neural network operation of a second teacher neural network to the training image to generate second pseudo label information, back-propagates a loss and updates parameters of the first and second student neural networks.

Abstract: A device is provided for adjusting a headrest for vehicles. The device includes a headrest unit connected to a vehicle seat and including a driving unit, a pair of side headrest units disposed on left and right sides of the headrest unit, respectively, and configured to be selectively rotated, and a pair of link units configured to respectively connect the pair of side headrest units to the headrest unit and to be operated by the driving unit to rotate the pair of side headrest units within a predetermined rotation adjustment range with respect to the headrest unit.

Abstract: A seat of a vehicle includes a foam pad in which a first hook part is installed and configured to slide on a rear surface opposite to a seating surface; a covering in which a wire is provided at an end portion and surrounds the seating surface of the foam pad to be caught and fixed to the first hook part of the rear surface of the foam pad; and a plate coupled to the rear surface of the foam pad and configured to slide the first hook part according to movement, wherein, when the covering and the foam pad are coupled, the hook part slides, and the wire is pulled by the first hook part so that the covering is fixed in a state of surrounding the foam pad.

Abstract: Disclosed is a device for adjusting an angle of a seatback for vehicles. The device includes a seat cushion unit including a locking part, a seatback unit rotatably connected to the seat cushion unit, and a link unit connecting the seat cushion unit to the seatback unit. The link unit includes an upper link part including an end fixed to the seat cushion unit to function as a rotation center axis and another end inserted into a back slot portion formed in the seatback unit, a lower link part including an end fixed to the seatback unit and another end inserted into a cushion slot portion formed in the seat cushion unit to be engaged with the locking part, and a joint part to which the upper and lower link parts are coupled. The seatback unit is rotated forward or backward by rotation of the upper and lower link parts.

Abstract: A vehicle foldable seat includes a seatback unit fixed to a vehicle body and coupled to a headrest and a seat cushion unit rotatably connected to the seatback unit. The seat cushion unit is fixed by elastic force in a stowing position, is selectively switched to a seating position when fixation thereof is released, and is engaged with the seatback unit to be fixed in the seating position.

Abstract: A seat for a vehicle, includes a second row center seat and a second row side seat provided on a partition wall positioned rearward of a driver seat, the second row center seat may move leftward or rightward, and an interval between the seats may be increased in a state in which the second row center seat is moved in a right direction away from the second row side seat, which makes it possible to maximally prevent body contact between a passenger in the second row center seat and a passenger in the second row side seat.

Abstract: Disclosed is a cushion tip-up type seat for a vehicle. The cushion tip-up type seat for a vehicle is configured to perform a tip-up function of a cushion part, and to move a seat leftward and rightward to adjust an interval between left and right seats, whereby left and right spacing between occupants seated in the seats is sufficiently secured and the convenience of the occupants is improved.

Abstract: The present disclosure relates to a vehicle rear seat including: a center seat; and side seats located on the left and right of the center seat, wherein, the center seat is capable of protruding by moving the center seat forward with respect to the side seats, and in the state in which the center seat protrudes forward, it is possible to increase an inter-passenger distance so that physical contact between the passenger of the center seat and the passenger of each of the side seats can be prevented as much as possible.