Abstract: Nowadays, demand for using real-time video transmission capable devices is ever-increasing. So, high resolution videos have made efficient video compression techniques an essential component for capturing and transmitting video data. Motion estimation has a critical role in encoding raw video. Hence, various motion estimation methods are introduced to efficiently compress the video. Low bit‑depth representation based motion estimation methods facilitate computation of matching criteria and thus, provide small hardware footprint. In this paper, a hardware implementation of a two-bit transformation based low-complexity motion estimation method using local binary pattern approach is proposed. Image frames are represented in two-bit depth instead of full-depth by making use of the local binary pattern as a binarization approach and the binarization part of the hardware architecture is explained in detail. Experimental results demonstrate the difference between the proposed hardware architecture and the architectures of well-known low-complexity motion estimation methods in terms of important aspects such as resource utilization, energy and power consumption.
Abstract: In this paper, an improved motion estimation (ME) approach based on weighted constrained one-bit transform is proposed for block-based ME employed in video encoders. Binary ME approaches utilize low bit-depth representation of the original image frames with a Boolean exclusive-OR based hardware efficient matching criterion to decrease computational burden of the ME stage. Weighted constrained one-bit transform (WC‑1BT) based approach improves the performance of conventional C-1BT based ME employing 2-bit depth constraint mask instead of a 1-bit depth mask. In this work, the range of constraint mask is further extended to increase ME performance of WC-1BT approach. Experiments reveal that the proposed method provides better ME accuracy compared existing similar ME methods in the literature.
Abstract: In this paper, an improved method for estimating fundamental matrix is proposed. The method is applied effectively to monocular camera based moving object detection. The method consists of corner points detection, moving object’s motion estimation and fundamental matrix calculation. The corner points are obtained by using Harris corner detector, motions of moving objects is calculated from pyramidal Lucas-Kanade optical flow algorithm. Through epipolar geometry analysis using RANSAC, the fundamental matrix is calculated. In this method, we have improved the performances of moving object detection by using two threshold values that determine inlier or outlier. Through the simulations, we compare the performances with varying the two threshold values.
Abstract: To acquire accurate ship motions at the center of gravity, a single low-cost inertial sensor is utilized and applied on board to measure ship oscillating motions. As observations, the three axes accelerations and three axes rotational rates provided by the sensor are used. The mathematical model of processing the observation data includes determination of the distance vector between the sensor and the center of gravity in x, y, and z directions. After setting up the transfer matrix from sensor’s own coordinate system to the ship’s body frame, an extended Kalman filter is applied to deal with nonlinearities between the ship motion in the body frame and the observation information in the sensor’s frame. As a side effect, the method eliminates sensor noise and other unwanted errors. Results are not only roll and pitch, but also linear motions, in particular heave and surge at the center of gravity. For testing, we resort to measurements recorded on a small vessel in a well-defined sea state. With response amplitude operators computed numerically by a commercial software (Seaway), motion characteristics are estimated. These agree well with the measurements after processing with the suggested method.
Abstract: Motion estimation occupies the heaviest computation in HEVC (high efficiency video coding). Many fast algorithms such as TZS (test zone search) have been proposed to reduce the computation. Still the huge computation of the motion estimation is a critical issue in the implementation of HEVC video codec. In this paper, motion estimator architecture with optimized number of PEs (processing element) is presented by exploiting early termination. It also reduces hardware size by exploiting parallel processing. The presented motion estimator architecture has 8 PEs, and it can efficiently perform TZS with very high utilization of PEs.
Abstract: Human action is recognized directly from the video sequences. The objective of this work is to recognize various human actions like run, jump, walk etc. Human action recognition requires some prior knowledge about actions namely, the motion estimation, foreground and background estimation. Region of interest (ROI) is extracted to identify the human in the frame. Then, optical flow technique is used to extract the motion vectors. Using the extracted features similarity measure based classification is done to recognize the action. From experimentations upon the Weizmann database, it is found that the proposed method offers a high accuracy.
Abstract: This paper integrates Octagon and Square Search
pattern (OCTSS) motion estimation algorithm into H.264/AVC
(Advanced Video Coding) video codec in Adaptive Group of Pictures
(AGOP) mode. AGOP structure is computed based on scene change
in the video sequence. Octagon and square search pattern block-based
motion estimation method is implemented in inter-prediction process
of H.264/AVC. Both these methods reduce bit rate and computational
complexity while maintaining the quality of the video sequence
respectively. Experiments are conducted for different types of video
sequence. The results substantially proved that the bit rate,
computation time and PSNR gain achieved by the proposed method
is better than the existing H.264/AVC with fixed GOP and AGOP.
With a marginal gain in quality of 0.28dB and average gain in bitrate
of 132.87kbps, the proposed method reduces the average computation
time by 27.31 minutes when compared to the existing state-of-art
H.264/AVC video codec.
Abstract: This paper presents an efficient VLSI architecture
design to achieve real time video processing using Full-Search Block
Matching (FSBM) algorithm. The design employs parallel bank
architecture with minimum latency, maximum throughput, and full
hardware utilization. We use nine parallel processors in our
architecture and each controlled by a state machine. State machine
control implementation makes the design very simple and cost
effective. The design is implemented using VHDL and the
programming techniques we incorporated makes the design
completely programmable in the sense that the search ranges and the
block sizes can be varied to suit any given requirements. The design
can operate at frequencies up to 36 MHz and it can function in QCIF
and CIF video resolution at 1.46 MHz and 5.86 MHz, respectively.
Abstract: SAD (Sum of Absolute Difference) algorithm is
heavily used in motion estimation which is computationally highly
demanding process in motion picture encoding. To enhance the
performance of motion picture encoding on a VLIW processor, an
efficient implementation of SAD algorithm on the VLIW processor is
essential. SAD algorithm is programmed as a nested loop with a
conditional branch. In VLIW processors, loop is usually optimized by
software pipelining, but researches on optimal scheduling of software
pipelining for nested loops, especially nested loops with conditional
branches are rare. In this paper, we propose an optimal scheduling and
implementation of SAD algorithm with conditional branch on a VLIW
DSP processor. The proposed optimal scheduling first transforms the
nested loop with conditional branch into a single loop with conditional
branch with consideration of full utilization of ILP capability of the
VLIW processor and realization of earlier escape from the loop. Next,
the proposed optimal scheduling applies a modulo scheduling
technique developed for single loop. Based on this optimal scheduling
strategy, optimal implementation of SAD algorithm on TMS320C67x,
a VLIW DSP is presented. Through experiments on TMS320C6713
DSK, it is shown that H.263 encoder with the proposed SAD
implementation performs better than other H.263 encoder with other
SAD implementations, and that the code size of the optimal SAD
implementation is small enough to be appropriate for embedded
environments.
Abstract: In this paper a new Joint Adaptive Block Matching
Search (JABMS) algorithm is proposed to generate motion vector
and search a best match macro block by classifying the motion vector
movement based on prediction error. Diamond Search (DS)
algorithm generates high estimation accuracy when motion vector is
small and Adaptive Rood Pattern Search (ARPS) algorithm can
handle large motion vector but is not very accurate. The proposed
JABMS algorithm which is capable of considering both small and
large motions gives improved estimation accuracy and the
computational cost is reduced by 15.2 times compared with
Exhaustive Search (ES) algorithm and is 1.3 times less compared
with Diamond search algorithm.
Abstract: Full search block matching algorithm is widely used for hardware implementation of motion estimators in video compression algorithms. In this paper we are proposing a new architecture, which consists of a 2D parallel processing unit and a 1D unit both working in parallel. The proposed architecture reduces both data access power and computational power which are the main causes of power consumption in integer motion estimation. It also completes the operations with nearly the same number of clock cycles as compared to a 2D systolic array architecture. In this work sum of absolute difference (SAD)-the most repeated operation in block matching, is calculated in two steps. The first step is to calculate the SAD for alternate rows by a 2D parallel unit. If the SAD calculated by the parallel unit is less than the stored minimum SAD, the SAD of the remaining rows is calculated by the 1D unit. Early termination, which stops avoidable computations has been achieved with the help of alternate rows method proposed in this paper and by finding a low initial SAD value based on motion vector prediction. Data reuse has been applied to the reference blocks in the same search area which significantly reduced the memory access.
Abstract: One of the most growing areas in the embedded community is multimedia devices. Multimedia devices incorporate a number of complicated functions for their operation, like motion estimation. A multitude of different implementations have been proposed to reduce motion estimation complexity, such as spiral search. We have studied the implementations of spiral search and identified areas of improvement. We propose a modified spiral search algorithm, with lower computational complexity compared to the original spiral search. We have implemented our algorithm on an embedded ARM based architecture, with custom memory hierarchy. The resulting system yields energy consumption reduction up to 64% and performance increase up to 77%, with a small penalty of 2.3 dB, in average, of video quality compared with the original spiral search algorithm.
Abstract: In this study, a classification-based video
super-resolution method using artificial neural network (ANN) is
proposed to enhance low-resolution (LR) to high-resolution (HR)
frames. The proposed method consists of four main steps:
classification, motion-trace volume collection, temporal adjustment,
and ANN prediction. A classifier is designed based on the edge
properties of a pixel in the LR frame to identify the spatial information.
To exploit the spatio-temporal information, a motion-trace volume is
collected using motion estimation, which can eliminate unfathomable
object motion in the LR frames. In addition, temporal lateral process is
employed for volume adjustment to reduce unnecessary temporal
features. Finally, ANN is applied to each class to learn the complicated
spatio-temporal relationship between LR and HR frames. Simulation
results show that the proposed method successfully improves both
peak signal-to-noise ratio and perceptual quality.
Abstract: In 3D-wavelet video coding framework temporal
filtering is done along the trajectory of motion using Motion
Compensated Temporal Filtering (MCTF). Hence computationally
efficient motion estimation technique is the need of MCTF. In this
paper a predictive technique is proposed in order to reduce the
computational complexity of the MCTF framework, by exploiting
the high correlation among the frames in a Group Of Picture (GOP).
The proposed technique applies coarse and fine searches of any fast
block based motion estimation, only to the first pair of frames in a
GOP. The generated motion vectors are supplied to the next
consecutive frames, even to subsequent temporal levels and only fine
search is carried out around those predicted motion vectors. Hence
coarse search is skipped for all the motion estimation in a GOP
except for the first pair of frames. The technique has been tested for
different fast block based motion estimation algorithms over different
standard test sequences using MC-EZBC, a state-of-the-art scalable
video coder. The simulation result reveals substantial reduction (i.e.
20.75% to 38.24%) in the number of search points during motion
estimation, without compromising the quality of the reconstructed
video compared to non-predictive techniques. Since the motion
vectors of all the pair of frames in a GOP except the first pair will
have value ±1 around the motion vectors of the previous pair of
frames, the number of bits required for motion vectors is also
reduced by 50%.
Abstract: In this paper, we consider the problem of tracking
multiple maneuvering targets using switching multiple target motion
models. With this paper, we aim to contribute in solving the problem
of model-based body motion estimation by using data coming from
visual sensors. The Interacting Multiple Model (IMM) algorithm is
specially designed to track accurately targets whose state and/or
measurement (assumed to be linear) models changes during motion
transition. However, when these models are nonlinear, the IMM
algorithm must be modified in order to guarantee an accurate track.
In this paper we propose to avoid the Extended Kalman filter because
of its limitations and substitute it with the Unscented Kalman filter
which seems to be more efficient especially according to the
simulation results obtained with the nonlinear IMM algorithm (IMMUKF).
To resolve the problem of data association, the JPDA
approach is combined with the IMM-UKF algorithm, the derived
algorithm is noted JPDA-IMM-UKF.
Abstract: In MPEG and H.26x standards, to eliminate the
temporal redundancy we use motion estimation. Given that the
motion estimation stage is very complex in terms of computational
effort, a hardware implementation on a re-configurable circuit is
crucial for the requirements of different real time multimedia
applications. In this paper, we present hardware architecture for
motion estimation based on "Full Search Block Matching" (FSBM)
algorithm. This architecture presents minimum latency, maximum
throughput, full utilization of hardware resources such as embedded
memory blocks, and combining both pipelining and parallel
processing techniques. Our design is described in VHDL language,
verified by simulation and implemented in a Stratix II
EP2S130F1020C4 FPGA circuit. The experiment result show that the
optimum operating clock frequency of the proposed design is 89MHz
which achieves 160M pixels/sec.
Abstract: In this paper a fast motion estimation method for
H.264/AVC named Triplet Search Motion Estimation (TS-ME) is
proposed. Similar to some of the traditional fast motion estimation
methods and their improved proposals which restrict the search points
only to some selected candidates to decrease the computation
complexity, proposed algorithm separate the motion search process to
several steps but with some new features. First, proposed algorithm try
to search the real motion area using proposed triplet patterns instead of
some selected search points to avoid dropping into the local minimum.
Then, in the localized motion area a novel 3-step motion search
algorithm is performed. Proposed search patterns are categorized into
three rings on the basis of the distance from the search center. These
three rings are adaptively selected by referencing the surrounding
motion vectors to early terminate the motion search process. On the
other hand, computation reduction for sub pixel motion search is also
discussed considering the appearance probability of the sub pixel
motion vector. From the simulation results, motion estimation speed
improved by a factor of up to 38 when using proposed algorithm than
that of the reference software of H.264/AVC with ignorable picture
quality loss.
Abstract: This paper introduces an intelligent system, which can be applied in the monitoring of vehicle speed using a single camera. The ability of motion tracking is extremely useful in many automation problems and the solution to this problem will open up many future applications. One of the most common problems in our daily life is the speed detection of vehicles on a highway. In this paper, a novel technique is developed to track multiple moving objects with their speeds being estimated using a sequence of video frames. Field test has been conducted to capture real-life data and the processed results were presented. Multiple object problems and noisy in data are also considered. Implementing this system in real-time is straightforward. The proposal can accurately evaluate the position and the orientation of moving objects in real-time. The transformations and calibration between the 2D image and the actual road are also considered.
Abstract: Recently, an enhanced hexagon-based search (EHS)
algorithm was proposed to speedup the original hexagon-based search
(HS) by exploiting the group-distortion information of some evaluated
points. In this paper, a second version of the EHS is proposed with a
new point-oriented inner search technique which can further speedup
the HS in both large and small motion environments. Experimental
results show that the enhanced hexagon-based search version-2
(EHS2) is faster than the HS up to 34% with negligible PSNR
degradation.
Abstract: In this paper, a fast motion compensation algorithm is
proposed that improves coding efficiency for video sequences with
brightness variations. We also propose a cross entropy measure
between histograms of two frames to detect brightness variations. The
framewise brightness variation parameters, a multiplier and an offset
field for image intensity, are estimated and compensated. Simulation
results show that the proposed method yields a higher peak signal to
noise ratio (PSNR) compared with the conventional method, with a
greatly reduced computational load, when the video scene contains
illumination changes.