A method for scale-aware depth estimation using multi-camera projection loss is described. The method includes determining a multi-camera photometric loss associated with a multi-camera rig of an ego vehicle. The method also includes training a scale-aware depth estimation model and an ego-motion estimation model according to the multi-camera photometric loss. The method further includes predicting a 360? point cloud of a scene surrounding the ego vehicle according to the scale-aware depth estimation model and the ego-motion estimation model. The method also includes planning a vehicle control action of the ego vehicle according to the 360? point cloud of the scene surrounding the ego vehicle.

A surround view system for a machine is provided. The surround view system includes a plurality of image capturing devices generating image data of surroundings of the machine, and an object detection system for detecting an object in a target field of view of the machine and generating object position data corresponding to the object. The surround view system also includes an image processing system configured to generate an initial 3-dimensional composite surround view by projecting the image data on a virtual model corresponding to the machine. The virtual model has a 3-dimensional shape based on an initial calibration position data. The image processing system is also configured to modify the 3-dimensional shape of the virtual model based on the object position data, and generate an updated 3-dimensional composite surround view by projecting the image data on the virtual model having a modified 3-dimensional shape.

An under vehicle image provision apparatus includes a plurality of bottom view cameras mounted to a bottom of a vehicle, a processor configured to generate an under vehicle image including tires of a vehicle based on images accessed from the bottom view cameras, and a display unit configured to display the under vehicle image. As such, an image showing the underside of the vehicle may be provided.

A vehicle surveillance system is disclosed and includes a plurality of image capturing units, an image processing unit, and a display unit for monitoring a position of at least one target around a vehicle and measuring a distance between the at least one target and the vehicle. The vehicle surveillance system utilizes a space domain determination module, a time domain determination module, and a ground surface elimination module to transform original images of the target around the vehicle into the bird's-eye-view panorama, and further detects variation of an optical pattern incident onto the target through a light beam emitted by a light source so as to effectively remind a driver on the on-going vehicle of the position of the target and the distance, and real-time detect and capture an image of any person approaching the vehicle, thereby achieving driving safety and securing lives and personal properties.

A vehicle display device includes: a display; a vehicle information image generator; an external image obtaining device; a display controller that displays the vehicle information image and the external image on the display; and an external image display determination device that determines whether the display controller is going to display the not displayed external image. When the display controller is going to display the not displayed external image, the display controller displays the external image, which the display controller is going to display, on the display with gradually increasing brightness of the external image from the brightness that is lower by a predetermined amount than original brightness to predetermined brightness that is determined based on the original brightness.

A pod includes a base, a plurality of sensor attachment fixtures, and a shell. The base is complementary in shape to a vehicle roof and includes a central opening therethrough. The sensor attachment fixtures are fixed to the base adjacent to an outer periphery of the base. The shell is fixed to and covers the base. The shell defines a cavity enclosing the sensor attachment fixtures and has a plurality of windows aligned with the sensor attachment fixtures.

Systems and methods for adjusting color and brightness in multi-camera systems are disclosed. The system uses four cameras, having overlapping fields of view, mounted on the four sides of a vehicle. Color errors are determined for areas where the images from the rear and side cameras' fields of view overlap. Color gain factors are determined and applied for each of the side cameras (using the rear camera as master) to match the colors of the video outputs from the side cameras and the rear camera. The gains for the front view camera are then adjusted using gain factors based on the matched video outputs from the side view cameras using the side cameras as the master to the front camera slave. In this way, the rear camera indirectly acts as the master for the front camera and all cameras are ultimately color-matched to the rear camera.

An image generation device for referencing a correspondence relationship and generating a line-of-sight-converted image from a captured image of an in-vehicle camera mounted to a vehicle is provided. The image generation device includes a first region updating unit that, upon sensing deviation of at least one of the mounting position and the mounting angle of the in-vehicle camera and calculating a new mounting position and mounting angle, updates a correspondence relationship of a predetermined first region in the line-of-sight-converted image in accordance with the new mounting position and mounting angle, and a second region updating unit that, upon satisfaction of a predetermined updating condition after updating the correspondence relationship of the first region, updates the correspondence relationship for a second region in the line-of-sight-converted image in accordance with the new mounting position and mounting angle.

Vehicle safety is enhanced by providing a video system incorporating multiple cameras providing separate video feeds that are stitched together by a controller to provide a composite video viewable by the operator that changes in real time along with changes to the speed and direction of the vehicle.

A vehicle three-dimensional image system includes a computing device configured to generate a synthesized three-dimensional image by acquiring a plurality of images from a plurality of cameras installed on a vehicle and arranging the plurality of images on a surface of a three-dimensional projection model, wherein the computing device is configured to generate the synthesized three-dimensional image by storing a plurality of three-dimensional projection models, selecting one of the plurality of three-dimensional projection models according to input information, and arranging the plurality of images on a surface of the selected three-dimensional projection model.