Signal processing apparatus and method of generating audio signal
A signal processing apparatus: specifies a capture time of a plurality of captured images that are captured from respectively different directions by a plurality of cameras in order to generate a virtual-viewpoint image corresponding to a designated virtual viewpoint; obtain an audio collection signal based on collection of audio of an audio collection target region by a microphone, wherein the audio collection target region includes at least a part of a capturing range of the plurality of cameras; generates an audio signal to be reproduced together with the virtual-viewpoint image corresponding to the capture time specified by the specification unit, by using the obtained audio collection signal based on collection by the microphone at a point in time prior to the specified capture time.
1. A signal processing apparatus, comprising:
at least one hardware processor; and
a memory which stores instructions executable by the at least one hardware processor to cause the signal processing apparatus to perform at least:
acquiring information related to a capturing point of time corresponding to a virtual-viewpoint image generated based on a plurality of captured images that are captured from respectively different directions by a plurality of cameras;
identifying a length of a playback period of the virtual-viewpoint image;
obtaining a collected sound signal based on collection of sound of a sound collection region by at least one microphone, wherein the sound collection region includes at least a part of a capturing area captured by the plurality of cameras; and
generating an audio signal to be reproduced together with the virtual-viewpoint image in the playback period, by using the obtained collected sound signal corresponding to a sound collection period determined based on both the capturing point of time specified by the acquired information and the identified length of the playback period,
wherein a sound collection point of time corresponding to the audio signal at a beginning of the playback period is earlier than the capturing point of time corresponding to the virtual-viewpoint image.
2. The signal processing apparatus according to claim 1 , wherein the instructions further cause the signal processing apparatus to perform:
executing control to cause the generated audio signal to be reproduced together with the virtual-viewpoint image.
3. The signal processing apparatus according to claim 1 , wherein the instructions further cause the signal processing apparatus to perform:
determining the sound collection period based on the capturing point of time specified by the acquired information and the identified length of the playback period.
4. The signal processing apparatus according to claim 1 , wherein the instructions further cause the signal processing apparatus to perform:
generating audio content which is reproduced together with image content including a period in which the virtual-viewpoint image is reproduced and a period in which a captured image captured by a camera is reproduced, the audio content including the generated audio signal.
5. The signal processing apparatus according to claim 1 , wherein
the virtual-viewpoint image is a moving image generated based on the plurality of captured images captured by the plurality of cameras at a single capturing point of time, and wherein
a virtual viewpoint corresponding to the virtual-viewpoint image changes over the playback period.
6. The signal processing apparatus according to claim 1 , wherein
the information related to the capturing point of time includes a time stamp of the virtual-viewpoint image.
7. The signal processing apparatus according to claim 4 , wherein, in a case where the period corresponding to the virtual-viewpoint image in the image content is from T 1 to T 2 , the audio signal which is to be reproduced in the period from T 1 to T 2 is generated by using a collected sound signal that is the same as the sound signal to be used to generate an audio signal to be reproduced in the period from 2T 1 −T 2 to T 1 .
8. The signal processing apparatus according to claim 1 , wherein the instructions further cause the signal processing apparatus to perform:
accepting input of image content, wherein
in a case where the accepted image content includes a virtual-viewpoint image which is a moving image corresponding to a single capturing point of time, the audio signal is generated by using a collected sound signal based on collection by the microphone at a point in time before the single capturing point of time.
9. The signal processing apparatus according to claim 4 , wherein the instructions further cause the signal processing apparatus to perform:
accepting input of image content, wherein
in a case where a capturing point of time corresponding to the captured image included in the accepted image content and a capturing point of time corresponding to the virtual-viewpoint image included in the image content are discontinuous, the audio signal is generated by using collected sound signal based on collection by the microphone at a point in time before the capturing point of time corresponding to the virtual-viewpoint image.
10. The signal processing apparatus according to claim 1 , wherein the instructions further cause the signal processing apparatus to perform:
accepting input of image content, wherein
in a case where a difference between a reproduction time of the accepted image content and a capturing point of time of a captured image to be used for generation of the image content changes within a period of the image content, the audio signal is generated by using a collected sound signal based on collection by the microphone at a point in time before the capturing point of time.
11. The signal processing apparatus according to claim 1 , wherein
a sound collection point of time corresponding to the audio signal at an end of the playback period is substantially the same as the capturing point of time corresponding to the virtual-viewpoint image.
12. The signal processing apparatus according to claim 1 , wherein the virtual-viewpoint image is a slow motion moving image.
13. The signal processing apparatus according to claim 1 , wherein
in the generating, the audio signal is generated by using the obtained collected sound signal based on collection by a microphone determined based on a position of a virtual viewpoint corresponding to the virtual-viewpoint image.
14. The signal processing apparatus according to claim 1 , wherein
in the generating, the audio signal is generated by using the obtained collected sound signal based on collection by a microphone determined based on a position of an object included in the plurality of captured images.
15. The signal processing apparatus according to claim 1 , wherein
the obtained collection sound signal includes both field audio in which a sound of a field of a stadium is and stand audio in which a sound of stands of the stadium and
in the generating the audio signal is generated by using the stand audio.
16. The signal processing apparatus according to claim 1 , wherein
in the generating, the audio signal corresponding to a collection position that changes in accordance with movement of a virtual viewpoint corresponding to the virtual-viewpoint image is generated.
17. The signal processing apparatus according to claim 1 , wherein
in the generating, the audio signal according to a position of a virtual viewpoint corresponding to the virtual-viewpoint image at the beginning of the playback period is generated.
18. The signal processing apparatus according to claim 1 , wherein
in the generating, the audio signal according to a position of a virtual viewpoint correspond to the virtual-viewpoint image at the end of the playback period is generated.
19. A method of generating an audio signal, comprising:
acquiring information related to a capturing point of time corresponding to a virtual-viewpoint image generate based on a plurality of captured images that are captured from respectively different directions by a plurality of cameras;
identifying a length of a playback period of the virtual-viewpoint image;
obtaining collected sound signal based on collection of a sound collection region by at least one microphone, wherein the sound collection region includes at least a part of a capturing area captured by the plurality of cameras; and
generating an audio signal to be reproduced together with the virtual-viewpoint image in the playback period, by using the obtained collected sound signal corresponding to a sound collection period determined based on both the capturing point of time specified by the acquired information and the identified length of the playback period, wherein a sound collection point of time corresponding to the audio signal at a beginning of the playback period is earlier than the capturing point of time corresponding to the virtual-viewpoint image.
20. A non-transitory computer-readable storage medium storing a program for causing a computer to execute a method of generating an audio signal, the method comprising:
acquiring information related to a capturing point of time corresponding to a virtual-viewpoint image generated based on a plurality of captured images that are captured from respectively different directions by a plurality of cameras;
obtaining collection collected sound signal based on collection of sound of a sound collection region by at least one microphone, wherein the sound collection region includes at least a part of a capturing area captured by the plurality of cameras; and
generating an audio signal to be reproduced together with the virtual-viewpoint image in the playback period, by using the obtained collected sound signal corresponding to a sound collection period determined based on both the capturing point of time specified by the acquired information and the identified length of the playback period, wherein a sound collection point of time corresponding to the audio signal at a beginning of the playback period is earlier than the capturing point of time corresponding to the virtual-viewpoint image.