Abstract: A model compression device includes a compression unit and a determination unit. The compression unit is configured to create a compression model arrived at by compressing a first prediction model created by machine learning. The determination unit is configured to determine whether or not a second prediction model created by re-learning the compression model can be further compressed on the basis of an index related to the performance of the second prediction model.

Abstract: Provided is an image processing device having an acquisition unit that acquires an image for learning used for machine learning; a target region setting unit that sets, to the image, a target region including a detection target; a detection region setting unit that sets, to the image, a detection region in which a teaching signal is required to be set; and a teaching signal setting unit that sets, to the detection region, a teaching signal that may take three or more values in accordance with a relevance between the detection region and the target region.

Abstract: Provided is an image processing device having an acquisition unit that acquires an image for learning used for machine learning; a target region setting unit that sets, to the image, a target region including a detection target; a detection region setting unit that sets, to the image, a detection region in which a teaching signal is required to be set; and a teaching signal setting unit that sets, to the detection region, a teaching signal that may take three or more values in accordance with a relevance between the detection region and the target region.

Abstract: A video processing apparatus includes a video generation unit 1, a load measurement unit 2, and a frame rate control unit 3. The video generation unit 1 generates respective videos with a frame rate of a predetermined setting value based on respective video signals of a plurality of imaging devices input from the plurality of imaging devices over a network. The load measurement unit 2 measures a load when the video generation unit 1 performs a process of generating the videos. The frame rate control unit 3 updates the predetermined setting value based on a measurement result of the load by the load measurement unit 2, and controls the video generation unit 1 to generate the respective videos at the frame rate of the updated predetermined setting value.