Showing items 1 - 20 of 160

Your selections:

A survey on evaluation of blockchain-based agricultural traceability

- Pang, Shaoning, Teng, Shyh, Murshed, Manzur, Bui, Cuong, Karmakar, Priyabrata, Li, Yanyu, Lin, Hao

  • Authors: Pang, Shaoning , Teng, Shyh , Murshed, Manzur , Bui, Cuong , Karmakar, Priyabrata , Li, Yanyu , Lin, Hao
  • Date: 2024
  • Type: Text , Journal article , Review
  • Relation: Computers and Electronics in Agriculture Vol. 227, no. (2024), p.
  • Full Text:
  • Reviewed:
  • Description: The integration of blockchain technology in agricultural traceability has shown immense potential, yet its widespread adoption faces significant roadblocks. Using bulk product traceability as a foundational reference, this paper presents a comprehensive evaluation framework for Blockchain-based Agricultural Traceability. The framework accentuates product identification and data traceability across the supply chain, addressing traceability disconnections caused by bulk product blending. It dives into depth levels from adoption decision-making to system design, development, and deployment, emphasizing the critical aspects of traceability compliance and standardization. As a result, we identified the obstacles to adopting agricultural digital traceability and pave the pathway to traceability system deployment. We examined the barriers to implementing digital traceability of agricultural products, taking the Australian grain supply chain as an example. Our findings reveal that lack of standardization and participation barriers are the primary challenges in implementing digital traceability for agricultural products. Our paper offers insights and recommendations for researchers, industry practitioners, and business owners to overcome these challenges and enable digital traceability of agricultural products in global supply chains. © 2024 The Authors

Automatic malleefowl mound detection using LiDAR-based ground and habitat features with planar terrain modelling

- Hossain, Nazia, Murshed, Manzur, Awrangjeb, Mohammad, Florentine, Singarayer, Irvin, Marc, Teng, Shyh

  • Authors: Hossain, Nazia , Murshed, Manzur , Awrangjeb, Mohammad , Florentine, Singarayer , Irvin, Marc , Teng, Shyh
  • Date: 2024
  • Type: Text , Conference paper
  • Relation: 2024 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2024, Niagra Falls, 15-19 July 2024, 2024 IEEE International Conference on Multimedia and Expo Workshops
  • Full Text: false
  • Reviewed:
  • Description: Over the recent years, airborne Light Detection and Ranging (LiDAR) data has been used to detect the egg incubators (a.k.a. mounds) of Malleefowl on large conservation areas. However, the mound detection accuracy of the existing automated methods is not high enough to be implemented in the field. To improve the detection accuracy, this paper focuses on the mound and its habitat-based feature extraction and classification, using the distinctive geometric behaviour of and surrounding the mound. We propose a novel feature extraction method of mounds, which leverages the slope of the ground with a planar terrain model (a.k.a. the ground plane) obtained by least squares plane fitting of the ground points. By observing the features based on the ground plane, we extract structural and habitat features, and group the mound-like ground points using clustering. Finally, we apply several binary classifier models to detect areas with mounds, among them Medium Gaussian SVM performs the best (around 95 % accuracy). Our training and testing datasets contain LiDAR point cloud data captured from the Tarawi Nature Reserve by the New South Wales Government in Australia. © 2024 IEEE.

BSDR : a data-efficient deep learning-based hyperspectral band selection algorithm using discrete relaxation

- Rahman, Mohammad, Teng, Shyh, Murshed, Manzur, Paul, Manoranjan, Brennan, David

  • Authors: Rahman, Mohammad , Teng, Shyh , Murshed, Manzur , Paul, Manoranjan , Brennan, David
  • Date: 2024
  • Type: Text , Journal article
  • Relation: Sensors Vol. 24, no. 23 (2024), p.
  • Full Text:
  • Reviewed:
  • Description: Hyperspectral band selection algorithms are crucial for processing high-dimensional data, which enables dimensionality reduction, improves data analysis, and enhances computational efficiency. Among these, attention-based algorithms have gained prominence by ranking bands based on their discriminative capability. However, they require a large number of model parameters, which increases the need for extensive training data. To address this challenge, we propose Band Selection through Discrete Relaxation (BSDR), a novel deep learning-based algorithm. BSDR reduces the number of learnable parameters by focusing solely on the target bands, which are typically far fewer than the original bands, thus resulting in a data-efficient configuration that minimizes training data requirements and reduces training time. The algorithm employs discrete relaxation, transforming the discrete problem of band selection into a continuous optimization task, which enables gradient-based search across the spectral dimension. Through extensive evaluations on three benchmark datasets with varying spectral dimensions and characteristics, BSDR demonstrates superior performance for both regression and classification tasks, achieving up to 25% and 34.6% improvements in overall accuracy, compared to the latest attention-based and traditional algorithms, respectively, while reducing execution time by 96.8% and 97.18%. These findings highlight BSDR’s effectiveness in addressing key challenges in hyperspectral band selection. © 2024 by the authors.

Crop monitoring by multimodal remote sensing : a review

- Karmakar, Priyabrata, Teng, Shyh, Murshed, Manzur, Pang, Shaoning, Li, Yanyu, Lin, Hao

  • Authors: Karmakar, Priyabrata , Teng, Shyh , Murshed, Manzur , Pang, Shaoning , Li, Yanyu , Lin, Hao
  • Date: 2024
  • Type: Text , Journal article , Review
  • Relation: Remote Sensing Applications: Society and Environment Vol. 33, no. (2024), p.
  • Full Text:
  • Reviewed:
  • Description: Effective approaches to achieve food safety and security can prevent catastrophic situations. Therefore, it is required to monitor agricultural crops on a regular basis. This can be easily achieved by capturing data from various remote sensing (RS) devices followed by processing them. Most RS devices are useful in monitoring crops and analysing different stages of plant growth successfully. However, individual devices have some limitations. To overcome this, multimodal remote sensing (MRS) methods have been gradually gaining popularity. In the multimodal approach, data from more than one modality are used together to obtain a better outcome. This is because, different modalities of data when used together can complement each other to achieve the same objective by combining their strengths and reducing their limitations, simultaneously. MRS methods have been found to be particularly useful for crop monitoring as they allow for the integration of data from multiple sources, resulting in a more comprehensive understanding of plant growth and development. By using MRS methods, it is possible to obtain a more accurate and detailed analysis of crop conditions, leading to improved decision-making and ultimately, better crop yields. In this paper, we will explore how MRS methods have been successfully utilised in crop monitoring and how the data obtained from these methods can provide valuable insights into the health and development of plants. © 2023 The Authors

Efficient motion modelling with variable-sized blocks from hierarchical cuboidal partitioning

- Karmakar, Priyabrata, Murshed, Manzur, Paul, Manoranjan, Taubman, David

  • Authors: Karmakar, Priyabrata , Murshed, Manzur , Paul, Manoranjan , Taubman, David
  • Date: 2024
  • Type: Text , Journal article
  • Relation: Multimedia Tools and Applications Vol. 83, no. 7 (2024), p. 20743-20757
  • Relation: http://purl.org/au-research/grants/arc/DP190102574
  • Full Text:
  • Reviewed:
  • Description: This paper explores the potential of cuboidal partitioning in motion modelling compared to the commonly used fixed-sized block-based architecture in scalable video coding. The traditional approach of dividing frames into fixed-sized blocks for independent motion compensation often results in coding inefficiency due to poor alignment with object boundaries. Hierarchical block partitioning has been introduced as a solution, but it suffers from an increased number of motion vectors, limiting its effectiveness. In contrast, cuboidal partitioning offers a promising alternative. It involves approximate segmentation of images into variable-sized rectangular segments (cuboids) that align well with object boundaries. The segmentation is based on a homogeneity constraint, minimizing the sum of squared errors (SSE). This property makes cuboidal partitioning compatible with block-based video coding techniques. In this paper, we investigate the potential of cuboids in motion modelling, specifically comparing them to fixed-sized blocks used in scalable video coding. Our approach involves constructing a motion-compensated current frame using the cuboidal partitioning information from the anchor frame within a group-of-pictures (GOP). The predicted current frame serves as the base layer, while the current frame is encoded as an enhancement layer using the scalable High Efficiency Video Coding (HEVC) encoder. Experimental results demonstrate significant bitrate savings ranging from 6.71% to 10.90% on 4 K video sequences. These savings highlight the superiority of our proposed model, which leverages cuboidal partitioning to improve coding efficiency and alignment with object boundaries. By adopting this approach, we mitigate the limitations of fixed-sized blocks and offer a more effective solution for motion modelling in scalable video coding. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023.

Machine learning accelerated prediction of 3D granular flows in hoppers

- Le, Duy, Nguyen, Linh, Phung, Truong, Howard, David, Kahandawa, Gayan, Murshed, Manzur, Delaney, Gary

Online state-to-state time-optimal trajectory planning for quadrotors in unknown cluttered environments

- Nguyen, Binh, Murshed, Manzur, Choudhury, Tanveer, Keogh, Kathleen, Appuhamillage, Gayan, Nguyen, Linh

  • Authors: Nguyen, Binh , Murshed, Manzur , Choudhury, Tanveer , Keogh, Kathleen , Appuhamillage, Gayan , Nguyen, Linh
  • Date: 2024
  • Type: Text , Conference paper
  • Relation: 2024 International Conference on Unmanned Aircraft Systems, ICUAS 2024, Chania, Crete, 4-7 June 2024, 2024 International Conference on Unmanned Aircraft Systems, ICUAS 2024 p. 309-316
  • Full Text: false
  • Reviewed:
  • Description: This paper introduces the first planner, called STAMINER (STAte-to-state tiMe-optImal memoryless planNER), which is able to real-time plan collision-free, local time-optimal trajectories in unknown and cluttered settings without the need for any maps or fused occupancy structures of the surrounding environment. Specifically, our method explores a library of state-to-state time-optimal trajectories in a memoryless collision-checking framework. It iteratively searches for the best trajectory with the longest projection on the direction to the goal. Results obtained from two different simulated cluttered scenarios demonstrate that our planner outperforms the state-of-the-art baselines concerning global finishing time. Furthermore, real-world flight trials were conducted to validate the effectiveness of our algorithm in an actual quadrotor. Finally, this paper also provides a mathematical proof of the solution characterization that is not available in the literature for the considered state-to-state time-optimal trajectory generation problem. © 2024 IEEE.

Unveiling soil-vegetation interactions : reflection relationships and an attention-based deep learning approach for carbon estimation

- Datta, Dristi, Paul, Manoranjan, Murshed, Manzur, Teng, Shyh, Schmidtke, Leigh

  • Authors: Datta, Dristi , Paul, Manoranjan , Murshed, Manzur , Teng, Shyh , Schmidtke, Leigh
  • Date: 2024
  • Type: Text , Conference paper
  • Relation: 2024 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2024, Niagra Falls, 15-19 July 2024, 2024 IEEE International Conference on Multimedia and Expo Workshops
  • Full Text: false
  • Reviewed:
  • Description: Estimating soil organic carbon (SOC) from satellite imagery, particularly in areas with both bare soil and vegetation, poses significant challenges. Traditional approaches often overlook the complex interactions between soil and vegetation. Addressing this gap, our study introduces an innovative method that leverages novel correction of hyperspectral reflections to adjust for vegetation levels, enhancing SOC estimation accuracy. Moreover, we propose an attention-based deep neural network that dynamically prioritizes spectral features crucial for SOC prediction. This mechanism significantly improves the model's ability to detect significant features for accurate SOC estimation. Comparative experiments with traditional models on a benchmark dataset demonstrate our method's effectiveness in reducing vegetation influence and accurately estimating SOC across mixed landscapes. Our findings represent a notable advancement in SOC estimation from satellite imagery, highlighting the potential of advanced learning-based techniques with attention-driven feature weighting for SOC estimation. © 2024 IEEE.

A depth-based hybrid approach for safe flight corridor generation in memoryless planning

- Nguyen, Thai, Murshed, Manzur, Choudhury, Tanveer, Keogh, Kathleen, Appuhamillage, Gayan, Nguyen, Linh

  • Authors: Nguyen, Thai , Murshed, Manzur , Choudhury, Tanveer , Keogh, Kathleen , Appuhamillage, Gayan , Nguyen, Linh
  • Date: 2023
  • Type: Text , Journal article
  • Relation: Sensors Vol. 23, no. 16 (2023), p.
  • Full Text:
  • Reviewed:
  • Description: This paper presents a depth-based hybrid method to generate safe flight corridors for a memoryless local navigation planner. It is first proposed to use raw depth images as inputs in the learning-based object-detection engine with no requirement for map fusion. We then employ an object-detection network to directly predict the base of polyhedral safe corridors in a new raw depth image. Furthermore, we apply a verification procedure to eliminate any false predictions so that the resulting collision-free corridors are guaranteed. More importantly, the proposed mechanism helps produce separate safe corridors with minimal overlap that are suitable to be used as space boundaries for path planning. The average intersection of union (IoU) of corridors obtained by the proposed algorithm is less than 2%. To evaluate the effectiveness of our method, we incorporated it into a memoryless planner with a straight-line path-planning algorithm. We then tested the entire system in both synthetic and real-world obstacle-dense environments. The obtained results with very high success rates demonstrate that the proposed approach is highly capable of producing safe corridors for memoryless local planning. © 2023 by the authors.

A robust local texture descriptor in the parametric space of the weibull distribution

- Tania, Sheikh, Karmakar, Gour, Teng, Shyh, Murshed, Manzur

  • Authors: Tania, Sheikh , Karmakar, Gour , Teng, Shyh , Murshed, Manzur
  • Date: 2023
  • Type: Text , Journal article
  • Relation: IEEE Transactions on Multimedia Vol. 25, no. (2023), p. 6053-6066
  • Full Text: false
  • Reviewed:
  • Description: Research in texture feature approximation is still in the embryonic stage because of difficulties in developing a sound theoretical model to express the unique pattern in the intensity-variation of pixels in the neighbourhood of the pixel-of-interest so that it can sufficiently discriminate different textures. Local texture descriptors are widely used in image segmentation as they comprise pixel-wise features. The Weber local descriptor (WLD) with differential excitation and gradient orientation components, inspired by Weber's Law, has been leveraged in the state-of-the-art iterative contraction and merging (ICM) image segmentation technique. However, WLD has inherent drawbacks in the formulation of the components that limit its discriminatory capability. This paper introduces a novel texture descriptor by directly modelling the distribution of intensity-variation in the parametric space of the Weibull distribution using its shape and scale parameters. A unified 'joint scale' texture property is introduced, which can discriminate textures better than the individual parameters while keeping the length of the descriptor shorter. Additionally, the accuracy of WLD's gradient orientation component is improved by using an extended Sobel operator and expressing gradients in -

Comparative analysis of machine and deep learning models for soil properties prediction from hyperspectral visual band

- Datta, Dristi, Paul, Manoranjan, Murshed, Manzur, Teng, Shyh, Schmidtke, Leigh

  • Authors: Datta, Dristi , Paul, Manoranjan , Murshed, Manzur , Teng, Shyh , Schmidtke, Leigh
  • Date: 2023
  • Type: Text , Journal article
  • Relation: Environments Vol. 10, no. 5 (2023), p. 77
  • Full Text:
  • Reviewed:
  • Description: Estimating various properties of soil, including moisture, carbon, and nitrogen, is crucial for studying their correlation with plant health and food production. However, conventional methods such as oven-drying and chemical analysis are laborious, expensive, and only feasible for a limited land area. With the advent of remote sensing technologies like multi/hyperspectral imaging, it is now possible to predict soil properties non-invasive and cost-effectively for a large expanse of bare land. Recent research shows the possibility of predicting those soil contents from a wide range of hyperspectral data using good prediction algorithms. However, these kinds of hyperspectral sensors are expensive and not widely available. Therefore, this paper investigates different machine and deep learning techniques to predict soil nutrient properties using only the red (R), green (G), and blue (B) bands data to propose a suitable machine/deep learning model that can be used as a rapid soil test. Another objective of this research is to observe and compare the prediction accuracy in three cases i. hyperspectral band ii. full spectrum of the visual band, and iii. three-channel of RGB band and provide a guideline to the user on which spectrum information they should use to predict those soil properties. The outcome of this research helps to develop a mobile application that is easy to use for a quick soil test. This research also explores learning-based algorithms with significant feature combinations and their performance comparisons in predicting soil properties from visual band data. For this, we also explore the impact of dimensional reduction (i.e., principal component analysis) and transformations (i.e., empirical mode decomposition) of features. The results show that the proposed model can comparably predict the soil contents from the three-channel RGB data.

Depth-based sampling and steering constraints for memoryless local planners

- Nguyen, Binh, Nguyen, Linh, Choudhury, Tanveer, Keogh, Kathleen, Murshed, Manzur

  • Authors: Nguyen, Binh , Nguyen, Linh , Choudhury, Tanveer , Keogh, Kathleen , Murshed, Manzur
  • Date: 2023
  • Type: Text , Journal article
  • Relation: Journal of Intelligent and Robotic Systems: Theory and Applications Vol. 109, no. 3 (2023), p.
  • Full Text:
  • Reviewed:
  • Description: By utilizing only depth information, the paper introduces a novel two-stage planning approach that enhances computational efficiency and planning performances for memoryless local planners. First, a depth-based sampling technique is proposed to identify and eliminate a specific type of in-collision trajectories among sampled candidates. Specifically, all trajectories that have obscured endpoints are found through querying the depth values and will then be excluded from the sampled set, which can significantly reduce the computational workload required in collision checking. Subsequently, we apply a tailored local planning algorithm that employs a direction cost function and a depth-based steering mechanism to prevent the robot from being trapped in local minima. Our planning algorithm is theoretically proven to be complete in convex obstacle scenarios. To validate the effectiveness of our DEpth-based both Sampling and Steering (DESS) approaches, we conducted experiments in simulated environments where a quadrotor flew through cluttered regions with multiple various-sized obstacles. The experimental results show that DESS significantly reduces computation time in local planning compared to the uniform sampling method, resulting in the planned trajectory with a lower minimized cost. More importantly, our success rates for navigation to different destinations in testing scenarios are improved considerably compared to the fixed-yawing approach. © 2023, The Author(s).

A commonality modeling framework for enhanced video coding leveraging on the cuboidal partitioning based representation of frames

- Ahmmed, Ashek, Murshed, Manzur, Paul, Manoranjan, Taubman, David

  • Authors: Ahmmed, Ashek , Murshed, Manzur , Paul, Manoranjan , Taubman, David
  • Date: 2022
  • Type: Text , Journal article
  • Relation: IEEE Transactions on Multimedia Vol. 24, no. (2022), p. 4446-4457
  • Full Text: false
  • Reviewed:
  • Description: Video coding algorithms attempt to minimize the significant commonality that exists within a video sequence. Each new video coding standard contains tools that can perform this task more efficiently compared to its predecessors. Modern video coding systems are block-based wherein commonality modeling is carried out only from the perspective of the block that need be coded next. In this work, we argue for a commonality modeling approach that can provide a seamless blending between global and local homogeneity information. For this purpose, at first the frame that need be coded, is recursively partitioned into rectangular regions based on the homogeneity information of the entire frame. After that each obtained rectangular region's feature descriptor is taken to be the average value of all the pixels' intensities encompassing the region. In this way, the proposed approach generates a coarse representation of the current frame by minimizing both global and local commonality. This coarse frame is computationally simple and has a compact representation. It attempts to preserve important structural properties of the current frame which can be viewed subjectively as well as from improved rate-distortion performance of a reference scalable HEVC coder that employs the coarse frame as a reference frame for encoding the current frame. © 1999-2012 IEEE.

Bidirectional mapping coupled GAN for generalized zero-shot learning

- Shermin, Tasfia, Teng, Shyh, Sohel, Ferdous, Murshed, Manzur, Lu, Guojun

  • Authors: Shermin, Tasfia , Teng, Shyh , Sohel, Ferdous , Murshed, Manzur , Lu, Guojun
  • Date: 2022
  • Type: Text , Journal article
  • Relation: IEEE Transactions on Image Processing Vol. 31, no. (2022), p. 721-733
  • Full Text:
  • Reviewed:
  • Description: Bidirectional mapping-based generalized zero-shot learning (GZSL) methods rely on the quality of synthesized features to recognize seen and unseen data. Therefore, learning a joint distribution of seen-unseen classes and preserving the distinction between seen-unseen classes is crucial for GZSL methods. However, existing methods only learn the underlying distribution of seen data, although unseen class semantics are available in the GZSL problem setting. Most methods neglect retaining seen-unseen classes distinction and use the learned distribution to recognize seen and unseen data. Consequently, they do not perform well. In this work, we utilize the available unseen class semantics alongside seen class semantics and learn joint distribution through a strong visual-semantic coupling. We propose a bidirectional mapping coupled generative adversarial network (BMCoGAN) by extending the concept of the coupled generative adversarial network into a bidirectional mapping model. We further integrate a Wasserstein generative adversarial optimization to supervise the joint distribution learning. We design a loss optimization for retaining distinctive information of seen-unseen classes in the synthesized features and reducing bias towards seen classes, which pushes synthesized seen features towards real seen features and pulls synthesized unseen features away from real seen features. We evaluate BMCoGAN on benchmark datasets and demonstrate its superior performance against contemporary methods. © 1992-2012 IEEE.

Discrete cosine basis oriented motion modeling with cuboidal applicability regions for versatile video coding

- Ahmmed, Ashek, Hamidouche, Wassim, Lambert, Andrew, Pickering, Mark, Murshed, Manzur

  • Authors: Ahmmed, Ashek , Hamidouche, Wassim , Lambert, Andrew , Pickering, Mark , Murshed, Manzur
  • Date: 2022
  • Type: Text , Conference paper
  • Relation: 2022 Picture Coding Symposium, PCS 2022, San Jose, Costa Rica, 7-9 December 2022, 2022 Picture Coding Symposium, PCS 2022 - Proceedings p. 337-341
  • Full Text: false
  • Reviewed:
  • Description: The relentless expansion of video based applications is underpinned by video coding technologies. The latest video coding standard i.e. versatile video coding (VVC) can provide superior compression performance than its predecessors. In this regard, motion modeling plays a central role. Experimental results showed that the discrete cosine basis oriented motion model can describe complex motion better than an affine motion model, adopted in the VVC. Hence, in this paper we propose to augment the VVC motion modeling technique with a set of discrete cosine basis oriented motion models and the applicability region of each such motion model is determined by non-overlapping rectangular regions, known as cuboids. Experimental results show a bit rate savings of up to 2.37% is achievable with respect to a VVC reference. © 2022 IEEE.

Dynamic mesh commonality modeling using the cuboidal partitioning

- Ahmmed, Ashek, Paul, Manoranjan, Murshed, Manzur, Pickering, Mark

  • Authors: Ahmmed, Ashek , Paul, Manoranjan , Murshed, Manzur , Pickering, Mark
  • Date: 2022
  • Type: Text , Conference paper
  • Relation: 2022 IEEE International Conference on Visual Communications and Image Processing, VCIP 2022, Suzhou, China, 13-16 December 2022, 2022 IEEE International Conference on Visual Communications and Image Processing, VCIP 2022
  • Full Text: false
  • Reviewed:
  • Description: For 3D object representation, volumetric contents like meshes and point clouds provide suitable formats. However, a dynamic mesh sequence may require significantly large amount of data because it consists of information that varies with time. Hence, for the facilitation of storage and transmission of such content, efficient compression technologies are required. MPEG has started standardization activities aiming to develop a mesh compression standard that would be able to handle dynamic meshes with time varying connectivity information and time varying attribute maps. The attribute maps are features associated with the mesh surface and stored as 2D images/videos. In this paper, we propose to capture the commonality information in the dynamic mesh attribute maps using the cuboidal partitioning algorithm. This algorithm is capable of modeling both the global and local commonality within an image in a compact and computationally efficient way. Experimental results show that the proposed approach can outperform the anchor HEVC codec, suggested by MPEG to encode such sequences, with a bit rate savings of up to 3.66%. © 2022 IEEE.

Efficient scalable 360-degree video compression scheme using 3d cuboid partitioning

- Afsana, Fariha, Paul, Manoranjan, Murshed, Manzur, Taubman, David

  • Authors: Afsana, Fariha , Paul, Manoranjan , Murshed, Manzur , Taubman, David
  • Date: 2022
  • Type: Text , Conference paper
  • Relation: 29th IEEE International Conference on Image Processing, ICIP 2022 p. 996-1000
  • Full Text: false
  • Reviewed:
  • Description: Video coding techniques minimize spatial and temporal redundancies inherent in video sequences based on non-overlapping block-based image partitioning. Due to depending on the information from already encoded neighboring blocks, these algorithms lack efficient techniques to exploit the overall global redundancies. Compared to the traditional block-based coding, the cuboid coding (2D) framework has been proven to be a more effective method of image compression that exploits global redundancy by considering homogeneous pixel correlation within a frame. In this paper, we improved the idea of 2D cuboid coding to exploit both local and global redundancy from a video sequence by adopting a three-dimensional (3D) cuboid partitioning scheme for SHVC compression improvement of 360-degree videos. The proposed method considers a group of successive frames as a 3D cuboid and recursively partitions it into sub-3D cuboids where static information over a selected GOP share the same cuboid and moving regions share new cuboids with better-defined objects. All the 3D cuboids are then encoded to create a coarse representation of the video stream. Experiments indicate that the proposed framework significantly outperforms its relevant benchmarks, notably by 17.18% (average) in BD-Rate reduction and 0.82 dB in BD-PSNR gain with respect to the standard SHVC codec. © 2022 IEEE.

Efficient scalable UHD/360-video coding by exploiting common information with cuboid-based partitioning

- Afsana, Fariha, Paul, Manoranjan, Murshed, Manzur, Taubman, David

  • Authors: Afsana, Fariha , Paul, Manoranjan , Murshed, Manzur , Taubman, David
  • Date: 2022
  • Type: Text , Journal article
  • Relation: IEEE Transactions on Circuits and Systems for Video Technology Vol. 32, no. 6 (2022), p. 3961-3977
  • Relation: http://purl.org/au-research/grants/arc/DP190102574
  • Full Text: false
  • Reviewed:
  • Description: The scalable extension of High Efficiency Video Coding, SHVC can code Ultra High-Definition (UHD) video, including 360-degree video for various devices to serve a single bitstream with different display resolutions and qualities. To improve the SHVC compression efficiency, this paper proposes a novel intra and inter-frame coding scheme by first separating the common/visually important information and then applying cuboid-based variable size block partitioning and coding process for the common/visually important information in the base layer. In cuboid-based partitioning a video frame is partitioned into arbitrary shaped rectangular regions, known as cuboids, based on the distribution of relatively homogeneous pixel values. As the cuboid adopts a variable block partitioning based on the homogeneity of the data value, the partitioned blocks have better alignment with the object boundary. Moreover, in the cuboid coding process, only the partitioning tree information and a single value for each block need to be coded which takes lower number of bits and computational time compared to the traditional SHVC base layer. To verify the performance of the proposed method we embedded the proposed scheme as a base layer into the standard SHVC reference software and used several popular UHD/360-degree videos. The experimental results indicate that the proposed scalable coding strategy achieves an average of 14.04% BD-Rate reduction and 0.61 dB BD-PSNR gain for UHD/360-video compared to the operation points provided by an SHVC conforming encoder. © 1991-2012 IEEE.

Human pose based video compression via forward-referencing using deep learning

- Rajin, S.M. Ataul Karim, Murshed, Manzur, Paul, Manoranjan, Teng, Shyh, Ma, Jiangang

  • Authors: Rajin, S.M. Ataul Karim , Murshed, Manzur , Paul, Manoranjan , Teng, Shyh , Ma, Jiangang
  • Date: 2022
  • Type: Text , Conference paper
  • Relation: 2022 IEEE International Conference on Visual Communications and Image Processing, VCIP 2022, Suzhou, China,13-16 December 2022, 2022 IEEE International Conference on Visual Communications and Image Processing, VCIP 2022
  • Full Text: false
  • Reviewed:
  • Description: To exploit high temporal correlations in video frames of the same scene, the current frame is predicted from the already-encoded reference frames using block-based motion estimation and compensation techniques. While this approach can efficiently exploit the translation motion of the moving objects, it is susceptible to other types of affine motion and object occlusion/deocclusion. Recently, deep learning has been used to model the high-level structure of human pose in specific actions from short videos and then generate virtual frames in future time by predicting the pose using a generative adversarial network (GAN). Therefore, modelling the high-level structure of human pose is able to exploit semantic correlation by predicting human actions and determining its trajectory. Video surveillance applications will benefit as stored 'big' surveillance data can be compressed by estimating human pose trajectories and generating future frames through semantic correlation. This paper explores a new way of video coding by modelling human pose from the already-encoded frames and using the generated frame at the current time as an additional forward-referencing frame. It is expected that the proposed approach can overcome the limitations of the traditional backward-referencing frames by predicting the blocks containing the moving objects with lower residuals. Our experimental results show that the proposed approach can achieve on average up to 2.83 dB PSNR gain and 25.93% bitrate savings for high motion video sequences compared to standard video coding. © 2022 IEEE.

Integrated generalized zero-shot learning for fine-grained classification

- Shermin, Tasfia, Teng, Shyh, Sohel, Ferdous, Murshed, Manzur, Lu, Guojun

  • Authors: Shermin, Tasfia , Teng, Shyh , Sohel, Ferdous , Murshed, Manzur , Lu, Guojun
  • Date: 2022
  • Type: Text , Journal article
  • Relation: Pattern Recognition Vol. 122, no. (2022), p.
  • Full Text:
  • Reviewed:
  • Description: Embedding learning (EL) and feature synthesizing (FS) are two of the popular categories of fine-grained GZSL methods. EL or FS using global features cannot discriminate fine details in the absence of local features. On the other hand, EL or FS methods exploiting local features either neglect direct attribute guidance or global information. Consequently, neither method performs well. In this paper, we propose to explore global and direct attribute-supervised local visual features for both EL and FS categories in an integrated manner for fine-grained GZSL. The proposed integrated network has an EL sub-network and a FS sub-network. Consequently, the proposed integrated network can be tested in two ways. We propose a novel two-step dense attention mechanism to discover attribute-guided local visual features. We introduce new mutual learning between the sub-networks to exploit mutually beneficial information for optimization. Moreover, we propose to compute source-target class similarity based on mutual information and transfer-learn the target classes to reduce bias towards the source domain during testing. We demonstrate that our proposed method outperforms contemporary methods on benchmark datasets. © 2021 Elsevier Ltd