This paper presented an approach to hide secret speech information in code excited linear prediction (CELP)-based speech coding scheme by adopting the analysis-by-synthesis (ABS)-based algorithm of speech information ...This paper presented an approach to hide secret speech information in code excited linear prediction (CELP)-based speech coding scheme by adopting the analysis-by-synthesis (ABS)-based algorithm of speech information hiding and extracting for the purpose of secure speech communication. The secret speech is coded in 2.4 Kb/s mixed excitation linear prediction (MELP), which is embedded in CELP type public speech. The ABS algorithm adopts speech synthesizer in speech coder. Speech embedding and coding are synchronous, i.e. a fusion of speech information data of public and secret. The experiment of embedding 2.4 Kb/s MELP secret speech in G.728 scheme coded public speech transmitted via public switched telephone network (PSTN) shows that the proposed approach satisfies the requirements of information hiding, meets the secure communication speech quality constraints, and achieves high hiding capacity of average 3.2 Kb/s with an excellent speech quality and complicating speakers’ recognition.展开更多
The proposed secure communication approach adopts the proposed algorithm of Analysis-By- Synthesis （ABS） speech information hiding to establish a Secret Speech Subliminai Channel （SSSC） for speech secure communica...The proposed secure communication approach adopts the proposed algorithm of Analysis-By- Synthesis （ABS） speech information hiding to establish a Secret Speech Subliminai Channel （SSSC） for speech secure communication over PSTN （Public Switched Telephone Network）, and employs the algorithm of ABS speech information extracting to recovery the secret information, This approach is more reliable, covert and securable than traditional and chaotic secure communication.展开更多
This paper presents a novel approach for camera pose refinement based on neural radiance fields(NeRF)by introducing semantic feature consistency to enhance robustness.NeRF has been successfully applied to camera pose ...This paper presents a novel approach for camera pose refinement based on neural radiance fields(NeRF)by introducing semantic feature consistency to enhance robustness.NeRF has been successfully applied to camera pose estimation by inverting the rendering process given an observed RGB image and an initial pose estimate.However,previous methods only adopted photometric consistency for pose optimization,which is prone to be trapped in local minima.To address this problem,we introduce semantic feature consistency into the existing framework.Specifically,we utilize high-level features extracted from a convolutional neural network(CNN)pre-trained for image recognition,and maintain consistency of such features between observed and rendered images during the optimization procedure.Unlike the color values at each pixel,these features contain rich semantic information shared within local regions and can be more robust to appearance changes from different viewpoints.Since it is computationally expensive to render a full image with NeRF for feature extraction from CNN,we propose an efficient way to estimate the features of individually rendered pixels by projecting them to a nearby reference image and interpolating its feature maps.Extensive experiments show that our method greatly outperforms the baseline method on both synthetic objects and real-world large indoor scenes,increasing the accuracy of pose estimation by over 6.4%.展开更多
文摘This paper presented an approach to hide secret speech information in code excited linear prediction (CELP)-based speech coding scheme by adopting the analysis-by-synthesis (ABS)-based algorithm of speech information hiding and extracting for the purpose of secure speech communication. The secret speech is coded in 2.4 Kb/s mixed excitation linear prediction (MELP), which is embedded in CELP type public speech. The ABS algorithm adopts speech synthesizer in speech coder. Speech embedding and coding are synchronous, i.e. a fusion of speech information data of public and secret. The experiment of embedding 2.4 Kb/s MELP secret speech in G.728 scheme coded public speech transmitted via public switched telephone network (PSTN) shows that the proposed approach satisfies the requirements of information hiding, meets the secure communication speech quality constraints, and achieves high hiding capacity of average 3.2 Kb/s with an excellent speech quality and complicating speakers’ recognition.
基金Supported by the National High Technology Research and Development Program of China (863 Program) (No.2003AA142080, 2004AA775060)the National Natural Sicence Foundation of China (No.60203004)+1 种基金with additional support from the China Post-doctorial Research Foundation (2005-03)the Foundation of Tianjin Key Lab for Advanced Signal Processing(2005).
文摘The proposed secure communication approach adopts the proposed algorithm of Analysis-By- Synthesis （ABS） speech information hiding to establish a Secret Speech Subliminai Channel （SSSC） for speech secure communication over PSTN （Public Switched Telephone Network）, and employs the algorithm of ABS speech information extracting to recovery the secret information, This approach is more reliable, covert and securable than traditional and chaotic secure communication.
文摘This paper presents a novel approach for camera pose refinement based on neural radiance fields(NeRF)by introducing semantic feature consistency to enhance robustness.NeRF has been successfully applied to camera pose estimation by inverting the rendering process given an observed RGB image and an initial pose estimate.However,previous methods only adopted photometric consistency for pose optimization,which is prone to be trapped in local minima.To address this problem,we introduce semantic feature consistency into the existing framework.Specifically,we utilize high-level features extracted from a convolutional neural network(CNN)pre-trained for image recognition,and maintain consistency of such features between observed and rendered images during the optimization procedure.Unlike the color values at each pixel,these features contain rich semantic information shared within local regions and can be more robust to appearance changes from different viewpoints.Since it is computationally expensive to render a full image with NeRF for feature extraction from CNN,we propose an efficient way to estimate the features of individually rendered pixels by projecting them to a nearby reference image and interpolating its feature maps.Extensive experiments show that our method greatly outperforms the baseline method on both synthetic objects and real-world large indoor scenes,increasing the accuracy of pose estimation by over 6.4%.
