IP Library Granted Patent US 9,451,164
Granted Patent B2
US 9,451,164 · App. 14/405,478 · Granted Sep 20, 2016

Video stabilisation with deblurring

Inventors: Stephane Auberger (Noisy le Grand, FR); Antoine Chouly (Paris, FR)
Assignee: ST-ERICSSON SA
H04N5/23267G06T5/003H04N5/23254G06T2207/20201
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,451,164
App. No.
14/405,478
Granted
Sep 20, 2016
Kind
B2
Abstract

Methods of processing a frame in a video sequence of digital images are provided. The methods include determining a global motion vector for the frame relative to a previous frame in the sequence, deriving a jitter function from the global motion vector, the jitter function including an estimate of undesired motion of the frame relative to the previous frame, and determining whether the frame is blurred above a first predetermined threshold. If the frame is blurred above the first predetermined threshold, stabilizing the frame using the jitter function and applying a deblur function to the frame.

Claims (25)

1. A method of processing a frame in a video sequence of digital images, the method comprising:

determining a global motion vector for the frame relative to a previous frame in the video sequence;

deriving a jitter function from the global motion vector, the jitter function comprising an estimate of an undesired motion of the frame relative to the previous frame;

determining whether the frame is blurred above a first predetermined threshold; and

if the frame is blurred above the first predetermined threshold, stabilising the frame using the jitter function and applying a deblur function to the frame.

2. The method according to claim 1 , further comprising

determining whether a moving object is present in the frame, and

if the moving object is present in the frame, and the frame is determined to be blurred above the first predetermined threshold, stabilising the frame without applying the deblur function.

3. The method according to claim 2 , wherein if no moving object is present in the frame, and the frame is determined to be blurred above the first predetermined threshold, stabilising the frame and applying the deblur function.

4. The method according to claim 1 , further comprising if the frame is not blurred above the first predetermined threshold, stabilising the frame without applying the deblur function.

5. The method according to claim 1 , wherein the determining of whether the frame is blurred comprises applying a high-pass filter to the frame.

6. The method according to claim 5 , wherein the high-pass filter comprises a Laplacian filter.

7. The method according to claim 1 , further comprising:

repairing the frame if the frame is determined to be blurred above a second predetermined threshold, the second threshold indicating that the frame is more blurred than the first threshold.

8. The method according to claim 7 , wherein the repairing of the frame comprises determining whether the previous frame is blurred above the second threshold, and if the previous frame is not blurred above the second threshold, then the frame is replaced with the previous frame.

9. The method according to claim 7 , wherein the repairing of the frame comprises determining whether a subsequent frame is blurred above the second threshold, and if the subsequent frame is not blurred above the second threshold, then the frame is replaced with a temporal upconversion of the previous frame and the subsequent frame.

10. The method according to claim 1 , wherein operations of the method are repeated for substantially all frames in the sequence.

11. The method according to claim 1 , wherein a kernel size for the deblur function is proportional to the global motion vector determined for the frame.

12. A non-transitory computer readable storing medium storing computer readable code which, when implemented on a processor causes the processor to perform the method according to claim 1 .

13. A device for processing a frame in a video sequence of digital images, the device comprising: a processor configured to:

determine a global motion vector for the frame relative to a previous frame in the video sequence;

derive a jitter function from the global motion vector, the jitter function comprising an estimate of an undesired motion of the frame relative to the previous frame;

determine whether the frame is blurred above a first predetermined threshold; and

if the frame is blurred above the first predetermined threshold, to stabilize the frame using the jitter function and applying a deblur function to the frame.

14. The device according to claim 13 , wherein the device further comprises one of a digital still camera, a digital camcorder or a mobile telephone including a digital camera.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 7, 2019
From: OPTIS CIRCUIT TECHNOLOGY, LLC,
To: TELEFONAKTIEBOLAGET L M ERICSSON (PUBL)
Reel/Frame 048529/0510 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 5, 2019
From: ST-ERICSSON SA, EN LIQUIDATION
To: OPTIS CIRCUIT TECHNOLOGY, LLC,
Reel/Frame 048504/0519 →
STATUS CHANGE-ENTITY IN LIQUIDATION Recorded Feb 2, 2016
From: ST-ERICSSON SA
To: ST-ERICSSON SA, EN LIQUIDATION
Reel/Frame 037739/0493 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 4, 2014
From: AUBERGER, STEPHANE; CHOULY, ANTOINE
To: ST-ERICSSON SA
Reel/Frame 034375/0570 →
Priority Claims (1)
EP 12305736 · Jun 25, 2012 · regional
Continuity (2)
Provisional Application 61671259 · Jul 13, 2012
Related Publication 20150138379A1 · May 21, 2015