Abstract: Systems, apparatus, and methods for dynamic transcoding based on video complexity. Transcoding inherits many of the same design assumptions as conventional video encoding which is optimized for content delivery networks that encode-once-deliver-often. Consumer grade encoders often transcode media on an as-needed basis since they lack the processing capability and/or volume of decodes to defray the cost of expensive encoding in advance. This typically occurs downstream from the camera and may be performed by different devices, with different parameters, to suit a variety of different usages. Codec optimization strategy mismatches are likely to occur in consumer ecosystems that combine different devices with competing design criteria. Dynamic transcoding based on previous codec information can improve the performance and/or capabilities of post-processing environments. Disclosed solutions dynamically adapt transcoding parameters (e.g., target bitrate, etc.

Abstract: A constraint model that includes a representation of feasible viewing window placement within a source field of view of visual content may be generated by using a roll-pitch-yaw axes representation of viewing window placement and having a diagonal dimension of the viewing window that fit within vertical and horizonal dimensions of the source field of view. The constraint model may enable full horizon leveling of the visual content.

Abstract: An image capture apparatus including a forward housing, a rear housing, and a front heatsink. The forward housing is a forward heatsink. The rear housing is a rear heatsink. The front heatsink houses all or a portion of an integrated sensor and lens assembly (ISLA).

Abstract: Systems, apparatus, and methods for real-time guided encoding. In one exemplary embodiment, an image processing pipeline (IPP) is implemented within a system-on-a-chip (SoC) that includes multiple stages, ending with a codec. The codec compresses video obtained from the previous stages into a bitstream for storage within removable media (e.g., an SD card), or transport (over e.g., Wi-Fi, Ethernet, or similar network). While most hardware implementations of real-time encoding allocate bit rate based on a limited look-forward (or look-backward) of the data in the current pipeline stage, the exemplary IPP leverages real-time guidance that was collected during the previous stages of the pipeline.

Abstract: An image capture system that includes an image capture apparatus and at least one optical accessory. The image capture apparatus includes a body and at least one lens that is supported by the body so as to define a field-of-view for the image capture apparatus. The at least one optical accessory is configured to overlie the at least one lens and thereby shift the field-of-view outwardly away from the image capture apparatus so as to define at least one blind area that is configured such that the field-of-view is spaced from the body of the image capture apparatus.

Abstract: A method provides capturing successive video image frames with an imaging device to generate a video. Processing the successive video image frames to determine positions of multiple subjects with the successive video image frames. Obtaining information for the multiple subjects from one or more sensors physically attached to respective ones of the multiple subjects. Determining a region of interest containing the multiple subjects in the successive video image frames of the video. Obtaining additional video image frames that include the multiple subjects. Processing the additional video image frames so that the one or more multiple subjects are maintained within the additional video image frames. Identifying a single subject out of the multiple subjects within the additional video image frames. Providing live videos that include the single subject to a sharing service via an application in communication with the imaging device regarding the single subject.

Abstract: A system includes a barrel mount disposed within a body, and the barrel mount includes a central axis. The system includes a lens barrel secured within the barrel mount and aligned with the central axis. The system includes an actuator that adjusts a position of the lens barrel by moving the barrel mount along the central axis. The system includes an actuator locking mechanism securable over the barrel mount and/or the lens barrel that prevents movement of the actuator and/or barrel mount when applying a force along the central axis towards the actuator.

Abstract: An image capture device with a housing, a drainage microphone, processor, and drainage channel. The housing defining an audio depression. The drainage microphone coupled to the housing at a location of the audio depression and configured to capture audio. The processor coupled with a memory storing instructions that when executed cause the processor to apply a set of tuned beamforming parameters to the captured audio. The drainage channel within the housing to drain moisture from the drainage microphone. The drainage channel defining a channel entrance having a channel entrance width and a channel entrance height.

Abstract: A video may include a capture of a scene, such as a wide-field of view capture of the scene. Context of the video may be assessed and used to suggest framing of the video. The framing of the video may be presented within a user interface. A user may select, through the user interface, the framing to be used in generating presentation of the video.

Abstract: An accessory is attached to an image capture device to augment, change, support, supplement, or modify the operation of the image capture device. Attachment of the accessory to the image capture device is detected by the image capture device, and the image capture device provides information on the attachment of the accessory.

Abstract: A video including visual content may be captured by an image capture device in motion. Stabilization performance information for the visual content may be determined. The stabilization performance information may characterize an extent to which desired stabilization is able to be performed using the visual content. The stabilization for the visual content may be changed based on the stabilization performance information.

Abstract: Adaptive acquisition control including limited luminance motion blur reduction for image and video acquisition and processing includes obtaining a target gain value and a target exposure duration value by obtaining a current scene luminance value, obtaining a current maximum gain using the current scene luminance value, obtaining a current interpolated gain-exposure duration curve using the current maximum gain and a current angular speed value, and obtaining the target gain value and the target exposure duration value from the current interpolated gain-exposure duration curve in accordance with a current maximum exposure duration. Adaptive acquisition control including limited luminance motion blur reduction for image and video acquisition and processing includes obtaining a frame of a video in accordance with the target scene exposition data and outputting the video.

Abstract: Systems and methods are disclosed for image signal processing. For example, methods may include, receiving an image from an image sensor; storing a sequence of images captured after the image in a buffer; determining an orientation error between the orientation of the image sensor and an orientation setpoint during capture of the image; determining a rotation corresponding to the orientation error based on orientation estimates from the sequence of orientation estimates corresponding to the sequence of images; and invoking an electronic image stabilization module to correct the image to obtain a stabilized image, in which the electronic image stabilization module corrects the image for the rotation corresponding to the orientation error.

Abstract: A video may be captured by an image capture device in motion. A horizon-leveled view of the video may be generated by providing a punchout of the video. The punchout of the video may compensate for rotation of the image capture device during capture of the video. The placement of the punchout of the video may be changed based on different rotational positions of to provide a view in which a horizon depicted within the video is leveled.

Abstract: Systems, apparatus, and methods for selectively parsing a live stream for a connected device. GoPro® cameras capture 2 video streams for every recording: a main resolution video (MRV) and a frame aligned low-resolution video (LRV). Currently, the LRV is streamed to other devices for playback in real-time. Unfortunately, not all devices in the mobile ecosystem can keep up with these transfer and playback speeds of the LRV; thus, exemplary embodiments of the present disclosure pick frames from the LRV to provide an IDR-only version that can be replayed with minimal processing burden. As a related issue, operating codecs in parallel at different resolutions can result in misaligned frame structures since each codec independently determines its “group of pictures” (GOP), etc. The described techniques define a frame correspondence that can also be used to improve post-processing at best effort.