IP Library Granted Patent US 12,445,659
Granted Patent B2
US 12,445,659 · App. 18/248,023 · Granted Oct 14, 2025

Electronic method and device for decoding a data stream, and associated computer program

Inventors: Félix Henry (Saint-Grégoire, FR); Gordon Clare (Pacé, FR)
Assignee: ORANGE
H04N19/91H04N19/436
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 12,445,659
App. No.
18/248,023
Granted
Oct 14, 2025
Kind
B2
Abstract

A method for decoding a bit sequence includes applying, at the input of an artificial neural network, previously decoded values; as a result of the previously decoded values being applied, producing a context index at the output of the artificial neural network; and obtaining a new decoded value by applying a portion of the bit sequence to an entropic decoder parameterized in the context identified by the produced context index. An electronic decoding device and an associated computer program are also provided.

Claims (25)

1. A method for decoding a sequence of binary elements, the method comprising:

applying, at an input of an artificial neural network, at least one previously decoded value obtained by an entropy decoder that is distinct from the artificial neural network;

producing a context index at an output of said artificial neural network, as a result of the applying the previously decoded values; and

obtaining at least one new decoded value by applying part of the sequence of binary elements to the entropy decoder parameterized in a context identified by the produced context index,

wherein a process of entropy decoding by the entropy decoder is suspended as long as a new context index is not produced at the output of the artificial neural network.

2. The decoding method according to claim 1 , further comprising applying the new decoded value to the input of the artificial neural network to produce, at the output of the artificial neural network, data representative of an audio or video content.

3. The decoding method according to claim 1 , wherein the sequence of binary elements is included in a data stream further including a piece of information indicative of a set of contexts usable within the entropy decoder.

4. The decoding method according to claim 3 , wherein said piece of information is indicative of a number of contexts usable within the entropy decoder.

5. The decoding method according to claim 1 , further comprising initializing each context usable within the entropy decoder using a parameterization piece of data included in a data stream including the sequence of binary elements.

6. The decoding method according to claim 1 , wherein the artificial neural network is implemented by a processor,

the method further comprising configuring the processor as a function of data included in a data stream including the sequence of binary elements.

7. The decoding method according to claim 1 , wherein the artificial neural network is implemented by a parallel processor configured to perform in parallel, at a specific time, a plurality of same operations.

8. The decoding method according to claim 7 , wherein the entropy decoder is implemented by another processor that is distinct from the parallel processor.

9. A non-transitory computer-readable medium on which is stored a computer program comprising instructions executable by a processor and configured to implement the decoding method according to claim 1 when the instructions are executed by the processor.

10. An electronic decoding device to decode a sequence of binary elements, the electronic device comprising:

an entropy decoder configured to receive the sequence of binary elements as an input;

an artificial neural network configured to receive at least one previously-decoded value obtained by the entropy decoder as an input and to produce a context index as an output, the entropy decoder being distinct from the artificial neural network; and

control circuitry configured to parameterize the entropy decoder in a context identified by the produced context index to obtain at least one new decoded value at an output of the entropy decoder; and

a synchronization system configured to suspend an entropy decoding process by the entropy decoder as long as a new context index is not produced at the output of the artificial neural network.

11. The electronic decoding device according to claim 10 , further comprising a processor configured to implement the artificial neural network.

12. The electronic decoding device according to claim 11 , wherein the control circuitry is configured to configure the processor as a function of data included in a data stream including the sequence of binary elements.

13. The electronic decoding device according to claim 11 , further comprising another processor that is distinct from the processor, the other processor being configured to implement the entropy decoder.

14. The decoding method according to claim 2 , wherein the sequence of binary elements is included in a data stream further including a piece of information indicative of a set of contexts usable within the entropy decoder.

15. The electronic decoding device according to claim 10 , further comprising a processor configured to implement the artificial neural network.

16. The decoding method according to claim 1 , wherein the artificial neural network comprises a context determination artificial neural network and a decoding artificial neural network.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 2, 2024
From: FONDATION B-COM
To: ORANGE
Reel/Frame 069452/0506 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 5, 2023
From: HENRY, FÉLIX; CLARE, GORDON
To: FONDATION B-COM
Reel/Frame 065134/0053 →
Priority Claims (1)
FR 2010214 · Oct 6, 2020 · national
Continuity (1)
Related Publication 20230379506A1 · Nov 23, 2023
References Cited (23)
US 10963782B2 · Xiong · 2021 [cited by examiner]
US 11095894B2 · Jeon · 2021 [cited by examiner]
US 11250595B2 · Covell · 2022 [cited by examiner]
US 20220004844A1 · Haase · 2022 [cited by examiner]
CN 110728726A · 2020 [cited by examiner]
CN 113658130A · 2021 [cited by examiner]
EP 3624452 · 2020 [cited by applicant]
WO 2018218249 · 2018 [cited by applicant]
WO 2022013247 · 2022 [cited by applicant]
WO 2022013249 · 2022 [cited by applicant]
WO 2022069437 · 2022 [cited by applicant]
Gary J. Sullivan et al., “Overview of the High Efficiency Video Coding (HEVC)N Standard”, IEEE Transactions on Circuit and Systems for Video Technology, vol. 22, No. 12, Dec. 2012, pp. 1649-1668. (Year: 2012). [cited by examiner]
Simon Wiedemann et al., “DeepCABAC: Context-adaptive binary arithmetic coding for deep neural network compression”, Proceedings of the 36th International Conference on Machine Learning, Long Beach, California, 2019, 4 p… [cited by examiner]
Jian-Kang Wu “Content-Based Indexing of Multimedia Databases”, Transactions on Image Processing, vol. 9, No. 6, Nov./Dec. 1997: 1041-4347 (c) 1997 IEEE (Year: 1997). [cited by examiner]
International Search Report dated Feb. 4, 2022, for PCT/EP2021/077157, 7 pp. [cited by applicant]
Written Opinion of the ISA dated Feb. 4, 2022, for PCT/EP2021/077157, 7 pp. [cited by applicant]
Detlev Marpe et al., “Content-Based Adaptive Binary Arithmetic Coding in the H.264/AVC Video Compression Standard”, IEEE Transactions on Circuit and Systems for Video Technology, vol. 13, No. 7, Jul. 2003, 18 pp. [cited by applicant]
Gary J. Sullivan et al., “Overview of the High Efficiency Video Coding (HEVC)N Standard”, IEEE Transactions on Circuit and Systems for Video Technology, vol. 22, No. 12, Dec. 2012, pp. 1649-1668. [cited by applicant]
Guo Lu et al., “DVC: An End-to-end Deep Video Compression Framework”, Computer Vision Foundation (CVF), 2019, pp. 11006-11015. [cited by applicant]
Anonymous, International Telecommunication Union, ITU-T, H.265, Telecommunication Standardization Sector of ITU, Series H: Audiovisual and Multimedia Systems, Infrastructure of audiovisual services—Coding of moving vide… [cited by applicant]
Simon Wiedemann et al., “DeepCABAC: Context-adaptive binary arithmetic coding for deep neural network compression”, Proceedings of the 36 [cited by applicant]
Simon Wiedemann et al., “Compact and Computationally Efficient Representation of Deep Neural Networks”, IEEE Transactions on Neural Networks and Learning Systems, vol. 31, No. 3, Mar. 2020, pp. 772-785. [cited by applicant]
English translation of the Written Opinion of the International Searching Authority dated Feb. 4, 2022 for corresponding International Application No. PCT/EP2021/077157, filed Oct. 1, 2021. [cited by applicant]