Image extraction from complex scenes in digital video

A method for masking an object from a background portion in a digital video, the method comprising:

• receiving a first user input defining a first border region that includes a border between an object and a background portion of a first digital image, the first digital image being one of a sequence of digital images that define a digital video, the first border region further including only a part of the object and only a part of the background portion, the border consisting pixels that include contributions from the object and the background portion;
• receiving a second user input defining a second border region that includes the border in a second digital image, the second digital image being another one of the sequence of digital images, the second border region further including only a part of the object and only a part of the background portion;
• interpolating between the first and second border regions to define an intermediary border region for an image intermediary in the sequence to the first and second digital images;
• using the first, second, and intermediary border regions for masking the object from the background portion in the digital video;
• using the first, second, and intermediary border regions for extracting the object from the background portion in the digital video;
• and
• placing the masked and extracted object against a new background in another digital video.

Computer software tangibly embodied in a computer-readable medium or a propagated carrier signal, for masking a foreground portion of from a background portion of a digital video, the software comprising instructions to perform the following operations:

• receive a first input defining a first border region that includes a border between a foreground portion and a background portion of a first digital image, the first digital image being one of a sequence of digital images that define a digital video, the first border region further including only a part of the foreground portion and only a part of the background portion, the border consisting of pixels that include contributions from the foreground portion and the background portion;
• receive a second input defining a second border region that includes the border in a second digital image, the second digital image being another one of the sequence of digital images, the second border region further including only a part of the foreground portion and only a part of the background portion;
• interpolate between the first and second border regions to define an intermediary border region for an image intermediary in the sequence to the first and second digital images;
• and
• use the first, second, and intermediary border regions for masking the foreground portion from the background portion in the digital video.

Computer software, tangibly embodied in a computer-readable medium or a propagated carrier signal, for masking an object from a background portion in a digital video, the software comprising instructions to perform the following operations:

• receive a first user input defining a first border region that includes a border between an object and a background portion of a first digital image, the first digital image being one of a sequence of digital images that define a digital video, the first border region further including only a part of the object and only a part of the background portion, the border consisting of pixels that include contributions from the object and the background portion;
• receive a second user input defining a second border region that includes the border in a second digital image, the second digital image being another one of the sequence of digital images, the second border region further including only a part of the object and only a part of the background portion;
• interpolate between the first and second border regions to define an intermediary border region for an image intermediary in the sequence to the first and second digital images;
• use the first, second, and intermediary border regions for masking the object from the background in the digital video;
• use the first, second, and intermediary border regions for extracting the object from the background portion in the digital video;
• and
• place the masked and extracted object against a new background in another digital video.