Spectral redundancy is a key element to be exploited in compression of remote sensing data. Combined with an entropy encoder, it can achieve competitive lossless coding performance. One of the latest techniques to dec...
详细信息
Spectral redundancy is a key element to be exploited in compression of remote sensing data. Combined with an entropy encoder, it can achieve competitive lossless coding performance. One of the latest techniques to decorrelate the spectral signal is the regression wavelet analysis (RWA). RWA applies a wavelet transform in the spectral domain and estimates the detail coefficients through the approximation coefficients using linear regression. RWA was originally coupled with JPEG 2000. This letter introduces a novel coding approach, where RWA is coupled with the predictor of CCSDS-123.0-B-1 standard and a lightweight contextual arithmetic coder. In addition, we also propose a smart strategy to select the number of RWA decomposition levels that maximize the coding performance. Experimental results indicate that, on average, the obtained coding gains vary between 0.1 and 1.35 bits-per-pixel-per-component compared with the other state-of-the-art coding techniques.
In predictive coding of integers real-valued residuals are mapped to integers before encoding, leaving room for improvement by reducing the loss due to rounding. In this paper, we propose a new prefix coding scheme wh...
详细信息
In predictive coding of integers real-valued residuals are mapped to integers before encoding, leaving room for improvement by reducing the loss due to rounding. In this paper, we propose a new prefix coding scheme where actual integer values, instead of the residuals, are encoded using cosets, with real domain predictions as the side information. This novel coding scheme outperforms Golomb-based coding by reducing the rounding loss with similar computational and memory complexity.
A novel image compression techinque is presented for low cost multimedia applications. The technique is based on quadtree segmented two-dimensional predictive coding for exploiting correlation between adjacent image b...
详细信息
A novel image compression techinque is presented for low cost multimedia applications. The technique is based on quadtree segmented two-dimensional predictive coding for exploiting correlation between adjacent image blocks and uniformity in variable block size image blocks. Low complexity visual pattern block truncation coding (VP-BTC) defined with a set of sixteen visual patterns is employed to code the high activity image blocks. Simulation results showed that the new technique achieved high performance with superior subjective quality at low bit rate.
The arithmetic coding of the bit planes of an image represented either in the conventional weighted binary form or the Gray code is investigated. Two-dimensional contexts based on the pixels from the current bit plane...
详细信息
The arithmetic coding of the bit planes of an image represented either in the conventional weighted binary form or the Gray code is investigated. Two-dimensional contexts based on the pixels from the current bit plane and three-dimensional contexts based on pixels from both the current and previous bit planes are studied. Results indicate that for binary bit planes, a three-dimensional template results in a significant bit rate reduction compared to a two-dimensional template of the same size. The savings are much less for Gray encoded bit planes;since the final bit rates are similar to the binary case, the need for Gray encoding is eliminated. The effect of the order of encoding the bit planes is also studied. Comparisons to conventional lossless schemes, such as predictive coding followed by Huffman or arithmetic coding, are presented.
The superior performance of tree encoding and vector quantization (vq) over scalar quantization (sq) is well known. However, large constraint length (in tree encoding) and large vector length (invq), which are require...
详细信息
The superior performance of tree encoding and vector quantization (vq) over scalar quantization (sq) is well known. However, large constraint length (in tree encoding) and large vector length (invq), which are required for a close-to-optimum performance, are limited by the exponentially growing complexity of the source encoder. The intention of our work has been to combine tree encoding andvq, each of moderate complexity, to preserve the advantages of both block and convolutional codes. This principle was applied to the prediction error signals of images as produced by non-adaptive and adaptive linear predictors.
predictive coding reflects the ability of the human brain to extract environmental patterns in order to reformulate previous expectations. The present report analyzes through the late N1 auditory component and the pos...
详细信息
predictive coding reflects the ability of the human brain to extract environmental patterns in order to reformulate previous expectations. The present report analyzes through the late N1 auditory component and the postimperative negative variation (PINV) the updating of predictions regarding the characteristics of a new trial, depending on the previous trial history, complexity, and type of trial (standard or deviant). Data were obtained from 31 healthy subjects recorded in a previous study based on two paradigms composed of stimulus sequences of decreasing or increasing frequencies intermingled with the sporadic appearance of unexpected tone endings. Our results showed a higher amplitude for the most complex condition and deviant trials for both the late N1 and PINV components. Additionally, the N1 and PINV presented a different amplitude response to the standard and deviant trials as a function of previous trial history, suggesting a continuous updating of trial categorization. The results suggest that the late N1 and PINV components are involved in the generation of an internal model about the rules of external auditory stimulation. NEW & NOTEWORTHY The present study showed a higher amplitude for the late N1 and the PINV with reference to both the prediction performed in auditory paradigms of higher abstraction and the unexpected breaking of the extracted rules. Additionally, both components were modulated depending on the local probability, which would suggest that they are continuously being updated by the previous stimulus history.
Repetition priming, that is, the repeated processing of a stimulus, facilitates performance. However, the neural underpinnings of repetition priming for famous faces in terms of effective connectivity are not known. H...
详细信息
Repetition priming, that is, the repeated processing of a stimulus, facilitates performance. However, the neural underpinnings of repetition priming for famous faces in terms of effective connectivity are not known. Here we investigated this problem using dynamic causal modelling of latency-corrected event-related brain potentials (RERPs). Source waveforms of RERP-derived sources in the Occipital Lobe, Fusiform Gyrus, Mediotemporal Lobe, Prefrontal Cortex and Anterotemporal Lobe of each hemisphere entered into models with only forward (F) or also with backward (FB) connections. Based on the framework of predictive coding formulated for repetition suppression, modulations of F and FB connections were expected for sources that displayed priming effects in their source waveforms. Hence, neural sources in each hemisphere were fitted with either F or FB connections. Inter-hemispheric connections were considered between homologous areas and were allowed to be modulated in an incremental manner resulting in a model space that comprised of 24 models. Bayesian model averaging across models revealed effective bidirectional connectivity between the Fusiform Gyrus (face perception) and Prefrontal Cortex (decision-making) in both hemispheres to be modulated by priming. In the left hemisphere, there is also a substantial involvement from the Mediotemporal Lobe, indicating the facilitation of automatic retrieval of the famous person's name. Furthermore, there is evidence that the priming is supported by connections from the right to the left Fusiform Gyri possibly in the service of inter-hemispheric cooperation. Altogether, the study indicates that along with top-down modulations, efficient processing within and across the two hemispheres is crucial for famous face priming.
A rare omission of a sound that is predictable by anticipatory visual information induces an early negative omission response (oN1) in the EEG during the period of silence where the sound was expected. It was previous...
详细信息
A rare omission of a sound that is predictable by anticipatory visual information induces an early negative omission response (oN1) in the EEG during the period of silence where the sound was expected. It was previously suggested that the oN1 was primarily driven by the identity of the anticipated sound. Here, we examined the role of temporal prediction in conjunction with identity prediction of the anticipated sound in the evocation of the auditory oN1. With incongruent audiovisual stimuli (a video of a handclap that is consistently combined with the sound of a car horn) we demonstrate in Experiment 1 that a natural match in identity between the visual and auditory stimulus is not required for inducing the oN1, and that the perceptual system can adapt predictions to unnatural stimulus events. In Experiment 2 we varied either the auditory onset (relative to the visual onset) or the identity of the sound across trials in order to hamper temporal and identity predictions. Relative to the natural stimulus with correct auditory timing and matching audiovisual identity, the oN1 was abolished when either the timing or the identity of the sound could not be predicted reliably from the video. Our study demonstrates the flexibility of the perceptual system in predictive processing (Experiment 1) and also shows that precise predictions of timing and content are both essential elements for inducing an oN1 (Experiment 2). (C) 2017 Elsevier B.V. All rights reserved.
The sight of a speaker's facial movements during the perception of a spoken message can benefit speech processing through online predictive mechanisms. Recent evidence suggests that these predictive mechanisms can...
详细信息
The sight of a speaker's facial movements during the perception of a spoken message can benefit speech processing through online predictive mechanisms. Recent evidence suggests that these predictive mechanisms can operate across sensory modalities, that is, vision and audition. However, to date, behavioral and electrophysiological demonstrations of cross-modal prediction in speech have considered only the speaker's native language. Here, we address a question of current debate, namely whether the level of representation involved in cross-modal prediction is phonological or pre-phonological. We do this by testing participants in an unfamiliar language. If cross-modal prediction is predominantly based on phonological representations tuned to the phonemic categories of the native language of the listener, then it should be more effective in the listener's native language than in an unfamiliar one. We tested Spanish and English native speakers in an audiovisual matching paradigm that allowed us to evaluate visual-to-auditory prediction, using sentences in the participant's native language and in an unfamiliar language. The benefits of cross-modal prediction were only seen in the native language, regardless of the particular language or participant's linguistic background. This pattern of results implies that cross-modal visual-to-auditory prediction during speech processing makes strong use of phonological representations, rather than low-level spatiotemporal correlations across facial movements and sounds.
We develop and evaluate motion compensation schemes for predictive coding of the component color television signal. Algorithms are discussed for estimation of motion of each color component (luminance and chrominance)...
详细信息
We develop and evaluate motion compensation schemes for predictive coding of the component color television signal. Algorithms are discussed for estimation of motion of each color component (luminance and chrominance) separately as well as in combination. Techniques for switching of the predictors for individual components are proposed and simulated. Simulations show that it is sufficient to estimate parameters of motion based only on the luminance and use them for motion-based prediction and switching the predictors for both the luminance and chrominance. Thus, only one motion estimator and prediction switch is needed for the three components of the color signal. Compression capability of motion compensation is scene dependent, in some video conference type of scenes, bit rate is reduced by as much as 60 percent compared to conditional replenishment coding.
暂无评论