Abstract: An apparatus includes an input circuit configured to receive a sequence of pictures and a processing circuit. The processing circuit may be configured to (i) remap image data of a first picture based upon a respective picture brightness values for the first picture and a second picture selected from the sequence of pictures, and (ii) perform temporal filtering between the first picture and the second picture utilizing the remapped image data.

Abstract: A method of rate-distortion computations for video compression is disclosed. The method may include steps (A) to (C). Step (A) may generate a plurality of transform coefficients from a residual block of the video using a circuit. Step (B) may generate a block distortion value (i) based on the transform coefficients and (ii) independent of a plurality of inverse transform samples produced from the residual block. Step (C) may generate a rate-distortion value from the block distortion value.

Abstract: A method for level-based motion detection in a sequence of pictures is disclosed. Step (A) of the method may compute one or more local tonal values in an area around a sample of a target picture in the sequence of pictures. Step (B) may compare the target picture and a reference picture in the sequence of pictures in the area to compute a motion score. Step (C) generally detects a motion of the sample based on the local tonal values, the motion score and one or more gain settings with a circuit.

Abstract: A method of pixel correction is disclosed. The method generally includes the steps of (A) calibrating a per-pixel correction model of a sensor at a plurality of different illumination levels, (B) generating a plurality of pixel values from the sensor in response to an optical signal and (C) generating a plurality of corrected values by applying the per-pixel correction model to the pixel values.

Abstract: A method for representing a plurality of data samples is disclosed. The method generally includes steps (A) and (B). Step (A) may generate an exponent common to all of the data samples using a circuit. Step (B) may generate a plurality of mantissae based on the exponent. Each of the mantissae generally corresponds to a different one of the data samples.

Abstract: A method for temporal filtering with local neighborhood limitations is disclosed. Step (A) of the method may perform a temporal filter operation on a target sample in a current picture of a sequence of pictures to generate an intermediate value of the target sample. Step (B) may find a minimum value and a maximum value among a plurality of local samples in a local neighborhood in the current picture around the target sample. The minimum value and the maximum value generally bound a minimum/maximum range. Step (C) may adjust the intermediate value toward the minimum/maximum range to generate a final value of the target sample if the intermediate value is outside a limited range. The limited range is generally based on the minimum/maximum range.

Abstract: An apparatus includes an input circuit configured to receive a sequence of pictures and a processing circuit. The processing circuit may be configured to (i) determine respective picture brightness values for each of a reference picture and a target picture selected from the sequence of pictures, (ii) remap image data of the reference picture based upon the respective picture brightness values, and (iii) perform temporal filtering between the reference picture and the target picture utilizing the remapped image data.

Abstract: An apparatus having a first circuit is disclosed. The first circuit may (i) receive an input signal from an electro-optical sensor and (ii) generate a compressed signal. The first circuit generally includes re-sampler circuits configured to process the input signal to generate output samples. Each output sample may be a linear combination of at least two input samples. Video processing may include at least two tasks among (a) generating a video signal by converting the input signal for an external video monitor, (b) generating a video signal by converting the input signal for a preview screen and (c) generating a video signal by resolution reduction of the input signal before compression. At least two re-sampler circuits may perform different tasks. Still processing generally includes a noise reduction of an input image of the input signal using at least one of the at least two re-sampler circuits.

Abstract: A method for processing a source frame is disclosed. The method generally includes the steps of (A) generating a intermediate video frame by de-mosaicing the source frame, wherein (i) the source frame has a source resolution and has a source color space and (ii) the intermediate video frame has the source resolution and has an intermediate color space, (B) generating a video stream by compressing the intermediate video frame with a video compression technique and (C) generating a still picture from the source frame, wherein (i) the still picture has the source resolution and comprises a raw picture and (ii) generating the still picture occurs substantially simultaneously with generating the video stream.

Abstract: A camera generally including a sensor and an auto-exposure circuit is disclosed. The sensor may be configured to generate a digital signal in response to an optical signal. The auto-exposure circuit may be configured to control a lightness of a picture within the digital signal by (i) adjusting at least one among an aperture, a shutter and an analog gain and (ii) adjusting a digital gain applied to the digital signal, wherein the digital gain adaptively reduces the lightness of the picture.

Abstract: A method for compressing a video stream including the steps of: (1) encoding at least two pictures of the video stream as a first intra picture and a second intra picture, where the first intra picture occurs temporally before the second intra picture, (2) encoding another picture of the video stream as a first inter picture, where the first inter picture (a) occurs temporally (i) after the first intra picture and (ii) before the second intra picture and (b) uses (i) the second intra picture and (ii) one other picture occurring temporally before the first inter picture as references, (3) encoding another picture of the video stream as a second inter picture, where the second inter picture (a) occurs temporally (i) after the first intra picture and (ii) before the second intra picture and (b) uses (i) the second intra picture and (ii) the first inter picture as references and (4) encoding another picture in the video stream as a third inter picture, where the third inter picture (a) occurs temporally after the

Abstract: A method for processing a digital picture is disclosed. The method may include steps (A) to (C). Step (A) may generate a first picture by processing the digital picture using a first noise reduction technique in a circuit. Step (B) may generate a second picture by processing the digital picture using a second noise reduction technique. The first noise reduction technique may achieve a higher noise reduction than the second noise reduction technique. Step (C) may generate an output picture by combining the first picture and the second picture.

Abstract: An apparatus generally comprising a first encoder, a decoder and a second encoder is disclosed. The first encoder may be configured to generate a plurality of first compressed samples from a plurality of data samples of a picture. The data samples generally include a plurality of luminance samples and chrominance samples created internal to a digital video camera. The decoder may be configured to generate a plurality of reconstructed samples from the first compressed samples. The second encoder may be configured to generate a plurality of second compressed samples based on the reconstructed samples.

Abstract: A method of pixel correction is disclosed. The method generally includes the steps of (A) calibrating a per-pixel correction model of a sensor at a plurality of different illumination levels, (B) generating a plurality of pixel values from the sensor in response to an optical signal and (C) generating a plurality of corrected values by applying the per-pixel correction model to the pixel values.

Abstract: An apparatus generally comprising a first encoder, a decoder and a second encoder is disclosed. The first encoder may be configured to generate a plurality of first compressed samples from a plurality of data samples of a picture. The data samples generally include a plurality of luminance samples and chrominance samples created internal to a digital video camera. The decoder may be configured to generate a plurality of reconstructed samples from the first compressed samples. The second encoder may be configured to generate a plurality of second compressed samples based on the reconstructed samples.

Abstract: An apparatus having a circuit is disclosed. The circuit may be configured to (i) process a digital image received from a camera sensor and (ii) convert the digital image after the processing. The converting generally uses a lookup table-based conversion that performs both (a) a color correction and (b) a tone correction.

Abstract: A method for compressing data samples using randomized quantization bins is disclosed. The method generally comprises steps (A) to (C). Step (A) may generate a size signal having a quantization step size corresponding to the data samples in an input signal using a compression circuit. The data samples generally reside in a range of numbers. Step (B) may generate the randomized quantization bins in the range based on a pseudorandom process. Step (C) may generate a plurality of quantized samples in a compressed signal by quantizing the data samples based on the randomized quantization bins.

Abstract: A method of capturing a still frame is disclosed. The method generally includes the steps of (A) generating a plurality of initial frames with a sensor in response to an optical signal and (B) generating the still frame by combining the initial frames using a noise reduction technique.

Abstract: A method for processing a source frame is disclosed. The method generally includes the steps of (A) generating a intermediate video frame by de-mosaicing the source frame, wherein (i) the source frame has a source resolution and has a source color space and (ii) the intermediate video frame has the source resolution and has an intermediate color space, (B) generating a video stream by compressing the intermediate video frame with a video compression technique and (C) generating a still picture from the source frame, wherein (i) the still picture has the source resolution and comprises a raw picture and (ii) generating the still picture occurs substantially simultaneously with generating the video stream.

Abstract: An apparatus having a circuit is disclosed. The circuit may be configured to (i) receive a digital image from an electro-optical sensor, (ii) convert the digital image from a red-green-blue representation to a luminance-and-chrominance representation, (iii) generate a reduced noise representation of the digital image by reducing noise in the luminance-and-chrominance representation and (iv) generate a color corrected representation of the digital image by color correcting the reduced noise representation.