Object detection

• US 20050129277A1
• Filed: 12/10/2004
• Published: 06/16/2005
• Est. Priority Date: 12/11/2003
• Status: Abandoned Application

• Current Assignee: Sony United Kingdom Limited (Sony Group Corp.)
• Original Assignee: Sony United Kingdom Limited (Sony Group Corp.)
• Inventors: Haynes, Simon Dominic, Living, Jonathan, Porter, Robert Mark Stefan, Rambaruth, Ratna
• Application Number:
  US11/009,842
• Publication Number:
  US 20050129277A1
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation
• CPC Class Codes: H04N 23/611 where the recognised objects include parts of the human body

Abstract

Object detection apparatus in which images from an ordered series of images are processed to detect objects, each image having an associated image period, the detection of objects in a whole image requiring a set of processing tasks to be carried out comprises means for executing subsets of the set of processing tasks during respective image periods associated with a group of two or more images from the ordered series; the subsets being arranged so that substantially all of the set of processing tasks are carried out in the course of the image periods associated with the group of images and that each such task is carried out with respect to at least one respective image in the group of images.

Claims

1. Object detection apparatus in which images from an ordered series of images are processed to detect objects, each image having an associated image period, the detection of objects in a whole image requiring a set of processing tasks to be carried out, said apparatus comprising:

logic to execute subsets of said set of processing tasks during respective image periods associated with a group of two or more images from said ordered series;

said subsets being arranged so that substantially all of said set of processing tasks are carried out in the course of the image periods associated with said group of images and that each such task is carried out with respect to at least one respective image in said group of images.

2. Apparatus according to claim 1, in which said series of images is a video signal.

3. Apparatus according to claim 1, in which said group of images is a contiguous group in said ordered series.

4. Apparatus according to claim 1, in which:

each image comprises an array of image elements; and

each subset of said set of processing tasks comprises processing tasks associated with a respective subset of said image elements.

5. Apparatus according to claim 4, in which, within a group of images, said subsets of image elements are non-overlapping.

6. Apparatus according to claim 4, in which, for a group having p images, a subset of image elements comprises every mth element in a horizontal direction and every nth element in a vertical direction, where n×

m=p.

7. Apparatus according to claim 6, in which said elements in a subset are offset on a row-by-row and/or a column-by-column basis.

8. Apparatus according to claim 4, in which said image elements are pixels.

9. Apparatus according to claim 4, in which said image elements are image blocks.

10. Apparatus according to claim 1, in which said subset of tasks carried out during each image period relate to the image associated with that image period.

11. Apparatus according to claim 10, in which said subsets are arranged so that an object detection result in respect of at least a part of each image is available after the subset of tasks relating to that image have been carried out.

12. Apparatus according to claim 1, in which said subsets are arranged so that substantially all of said set of processing tasks are carried out in the course of the image periods associated with said group of images and that each such task is carried out with respect to at least one respective image in a subgroup selected from said group of images.

13. Apparatus according to claim 12, in which said subgroup comprises a single image within said group.

14. Apparatus according to claim 13, in which said subsets are arranged so that object detection is carried out in respect of discrete spatial portions of said single image over respective numbers of image periods within said group.

15. Apparatus according to claim 1, comprising:

estimator logic to estimate the processing operations required by tasks in said set of processing tasks; and

allocator logic to allocate a subset of said set of processing tasks to images within a group so as not to exceed a processing capacity available in respect of each image.

16. Apparatus according to claim 1, in which:

said set of processing tasks is arranged as two or more chains of processing tasks; and

said allocator logic is operable to allocate tasks to images in said group as complete chains of tasks.

17. Apparatus according to claim 16, in which:

objects are detected using multiple sets of object probability data relating to different respective expected object sizes relative to the image size; and

said processing tasks required to handle one of said sets of probability data are treated as a respective chain.

18. Apparatus according to claim 1, in which said objects are faces.

19. Object tracking apparatus comprising:

object detection apparatus according claim 1;

and

logic to associate detected objects in images of said ordered series so as to track an individual object from image to image.

20. Object tracking apparatus according to claim 19, in which:

if, at an image period, said subsets of processing tasks completed so far are such that an object detection is not yet available in respect of a group, object tracking is not carried out at that image period.

21. Object tracking apparatus according to claim 20, in which, where said processing tasks in respect of object detection are arranged over a group of p images, object tracking is carried out after processing said group of p images.

22. Object tracking apparatus according to claim 19, comprising:

a second object detector for detecting the presence of object(s) in said images, said second object detector having a lower detection threshold than the object detection apparatus, so that said second object detector is more likely to detect an object in an image region in which said object detection apparatus has not detected an object;

in which;

if, during the image periods associated with a group of images, the completed subsets of processing tasks are such that an object detection is not yet available, object tracking is carried out using the results of said second object detector.

23. Object tracking apparatus according to claim 22, comprising:

an object position predictor for predicting an object position in a next image in a test order of the video sequence on the basis of a detected object position in one or more previous images in said test order of said video sequence;

in which;

if said apparatus detects an object within a predetermined threshold image distance of the predicted object position, said object position predictor uses said detected position to produce a next position prediction;

or

if said apparatus fails to detect an object within a predetermined threshold image distance of said predicted object position, said object position predictor uses an object position detected by said second object detector to produce a next position prediction.

24. Video conferencing apparatus comprising apparatus according to claim 1.

25. Surveillance apparatus comprising apparatus according to claim 1.

26. A method of object detection, in which a test image from an ordered series of images is processed to detect objects in that image, the detection of objects in a whole image requiring a set of processing tasks to be carried out, said method comprising the step of:

executing a subset of said set of processing tasks with respect to said test image, said subset being arranged as part of a sequence over a group including said test image and at least one other image in said ordered series so that substantially all of said set of processing tasks are carried out with respect to at least one image in said group.

27. Computer software having program code for carrying out a method according to claim 26.

28. A providing medium for providing program code according to claim 27.

29. A medium according to claim 28, said medium being a storage medium.

30. A medium according to claim 28, said medium being a transmission medium.