Conditional Variational Autoencoder Research Articles

Two-stage correction prediction of wind power based on numerical weather prediction wind speed superposition correction and improved clustering

The unavoidable wind power prediction error poses a challenge for wind power to participate in day-ahead scheduling plan. This paper proposes a two-stage correction method considering NWP wind speed and power correction. Firstly, this paper fully considers the historical similarity of NWP wind speed, and adopts the improved clustering distance for segment matching and wind speed correction based on the extracted most relevant trend component, which enhances the utilization of NWP information and avoids the randomness and uncertainty of the intelligent correction method, and gram angular field (GAF) and stacked autoencoder (SAE) are used for feature extraction and conditional variational autoencoder (CVA) is used to generate specific samples to ensure the adequacy of the clustering process. Secondly, considering the historical similarity of error of predicted power, the power is further corrected based on the proposed weighted double-constraint to realize the secondary calibration. The Proposed method is applied to several wind farms in China to verify its effectiveness. The results show that the proposed method reduces the NRMSE by 3.82 % and NMAE by 3.40 % compared with direct prediction in the wind farms in western Inner Mongolia, which is important for promoting wind power consumption and maintaining the safe and stable operation of the power system.

Expressway Vehicle Trajectory Prediction Based on Fusion Data of Trajectories and Maps from Vehicle Perspective

Research on vehicle trajectory prediction based on road monitoring video data often utilizes a global map as an input, disregarding the fact that drivers rely on the road structures observable from their own positions for path planning. This oversight reduces the accuracy of prediction. To address this, we propose the CVAE-VGAE model, a novel trajectory prediction approach. Initially, our method transforms global perspective map data into vehicle-centric map data, representing it through a graph structure. Subsequently, Variational Graph Auto-Encoders (VGAEs), an unsupervised learning framework tailored for graph-structured data, are employed to extract road environment features specific to each vehicle’s location from the map data. Finally, a prediction network based on the Conditional Variational Autoencoder (CVAE) structure is designed, which first predicts the driving endpoint and then fits the complete future trajectory. The proposed CVAE-VGAE model integrates a self-attention mechanism into its encoding and decoding modules to infer endpoint intent and incorporate road environment features for precise trajectory prediction. Through a series of ablation experiments, we demonstrate the efficacy of our method in enhancing vehicle trajectory prediction metrics. Furthermore, we compare our model with traditional and frontier approaches, highlighting significant improvements in prediction accuracy.

Creating synthetic energy meter data using conditional diffusion and building metadata

Advances in machine learning and increased computational power have driven progress in energy-related research. However, limited access to private energy data from buildings hinders traditional regression models relying on historical data. While generative models offer a solution, previous studies have primarily focused on short-term generation periods (e.g., daily profiles) and a limited number of meters. Thus, the study proposes a conditional diffusion model for generating high-quality synthetic energy data using relevant metadata. Using a dataset comprising 1,828 power meters from various buildings and countries, this model is compared with traditional methods like Conditional Generative Adversarial Networks (CGAN) and Conditional Variational Auto-Encoders (CVAE). It explicitly handles long-term annual consumption profiles, harnessing metadata such as location, weather, building, and meter type to produce coherent synthetic data that closely resembles real-world energy consumption patterns. The results demonstrate the proposed diffusion model's superior performance, with a 36% reduction in Fréchet Inception Distance (FID) score and a 13% decrease in Kullback-Leibler divergence (KL divergence) compared to the following best method. The proposed method successfully generates high-quality energy data through metadata, and its code will be open-sourced, establishing a foundation for a broader array of energy data generation models in the future.

Design Space Exploration and Explanation via Conditional Variational Autoencoders in Meta-Model-Based Conceptual Design of Pedestrian Bridges

Today, engineers rely on conventional iterative (often manual) techniques for conceptual design. Emerging parametric models facilitate design space exploration based on quantifiable performance metrics, yet remain time-consuming and computationally expensive, leaving room for improvement. This paper provides a design exploration and explanation framework to augment the designer via a Conditional Variational Autoencoder (CVAE), which serves as a forward performance predictor as well as an inverse design generator conditioned on a set of performance requests. Hence, the CVAE overcomes the limitations of traditional iterative techniques by learning a differentiable mapping for a highly nonlinear design space, thus enabling sensitivity analysis. These methods allow for informing designers about (i) relations of the model between features and performances and (ii) structural improvements under user-defined objectives. The framework is tested on a case-study and proves its potential to serve as a future co-pilot for conceptual design studies of diverse civil structures and beyond.

A deep learning-based pipeline for developing multi-rib shape generative model with populational percentiles or anthropometrics as predictors

Rib cross-sectional shapes (characterized by the outer contour and cortical bone thickness) affect the rib mechanical response under impact loading, thereby influence the rib injury pattern and risk. A statistical description of the rib shapes or their correlations to anthropometrics is a prerequisite to the development of numerical human body models representing target demographics. Variational autoencoders (VAE) as anatomical shape generators remain to be explored in terms of utilizing the latent vectors to control or interpret the representativeness of the generated results. In this paper, we propose a pipeline for developing a multi-rib cross-sectional shape generative model from CT images, which consists of the achievement of rib cross-sectional shape data from CT images using an anatomical indexing system and regular grids, and a unified framework to fit shape distributions and associate shapes to anthropometrics for different rib categories. Specifically, we collected CT images including 3193 ribs, surface regular grid is generated for each rib based on anatomical coordinates, the rib cross-sectional shapes are characterized by nodal coordinates and cortical bone thickness. The tensor structure of shape data based on regular grids enable the implementation of CNNs in the conditional variational autoencoder (CVAE). The CVAE is trained against an auxiliary classifier to decouple the low-dimensional representations of the inter- and intra- variations and fit each intra-variation by a Gaussian distribution simultaneously. Random tree regressors are further leveraged to associate each continuous intra-class space with the corresponding anthropometrics of the subjects, i.e., age, height and weight. As a result, with the rib class labels and the latent vectors sampled from Gaussian distributions or predicted from anthropometrics as the inputs, the decoder can generate valid rib cross-sectional shapes of given class labels (male/female, 2nd to 11th ribs) for arbitrary populational percentiles or specific age, height and weight, which paves the road for future biomedical and biomechanical studies considering the diversity of rib shapes across the population.

Metamaterial-based realization for thermal transparency: A conditional variational autoencoder approach

The advances in thermal metamaterials and their applications have revolutionized how we can manipulate thermal transport behavior. The challenging inverse design problems of utilizing thermal metamaterial-based structures to achieve desired thermal transport behavior are increasingly being tackled by data-driven, machine learning-based approaches. The explosive progress in generative AI is permeating the field of material design by offering new perspectives to address the inverse design problems. In this paper, we propose a simple yet effective method of training a generative conditional variational autoencoder to find the design parameters for a thermal metamaterial-based system with a periodic interparticle arrangement to achieve thermal transparency, which is one of the most desirable and interesting thermal transport behaviors. Our work attests to the predictive power of a generative model with a relatively small number of parameters for the purpose of tackling inverse design problems to achieve thermal transport behavior manipulation.

The covariance environment defines cellular niches for spatial inference.

A key challenge of analyzing data from high-resolution spatial profiling technologies is to suitably represent the features of cellular neighborhoods or niches. Here we introduce the covariance environment (COVET), a representation that leverages the gene-gene covariate structure across cells in the niche to capture the multivariate nature of cellular interactions within it. We define a principled optimal transport-based distance metric between COVET niches that scales to millions of cells. Using COVET to encode spatial context, we developed environmental variational inference (ENVI), a conditional variational autoencoder that jointly embeds spatial and single-cell RNA sequencing data into a latent space. ENVI includes two decoders: one to impute gene expression across the spatial modality and a second to project spatial information onto single-cell data. ENVI can confer spatial context to genomics data from single dissociated cells and outperforms alternatives for imputing gene expression on diverse spatial datasets.

Chiller fault diagnosis based on improved variational autoencoder and co-training framework: A case study of insufficient samples

Implementing efficient automatic fault diagnosis is critical for saving energy and minimizing financial losses in the heating ventilation air-conditioning (HVAC) systems of commercial buildings. However, the limited quantity and weak features of fault samples acquired during HVAC operations hinder the effectiveness of conventional machine learning-based fault diagnosis methodologies. This paper proposes a method based on an improved conditional variational autoencoder (MCVAE) and co-training (CT) ideology to address the issue of insufficient training samples. Initially, we employ MCVAE to synthesize an extensive dataset of chiller fault samples from the original training dataset. Subsequently, the beneficial samples for training our fault diagnostic classifier, namely high-quality samples, are selected from the generated dataset using the CT-based framework. Finally, the selected high-quality samples are merged into the original training dataset to train the ultimate fault classifiers. Experimental results demonstrate that our proposed method outperforms in effectiveness and efficiency compared to recently published methods. For instance, in the case of fault level 1 compared to the suboptimal model, our approach exhibits improvements of 2.41% when each type has 5 fault samples.

Indoor positioning fingerprint database construction based on CSA-DBSCAN and RCVAE-GAN

With the increasing size of buildings, in order to achieve high-precision indoor positioning services, it is a challenging task to build an offline fingerprint database with high quality, high density and less manpower and material consumption. Aiming to solve the problem of low-quality WiFi indoor positioning fingerprint inventory constructed by traditional methods, which affects positioning accuracy and incurs high costs, this paper proposes a method for indoor positioning fingerprint database construction based on Crow Search Algorithm Optimizes Density Clustering (CSA-DBSCAN) and Regressor Conditional VAE Generative Adversarial Network (RCVAE-GAN). Collecting only a tiny amount of sparse reference point position coordinates and RSS data makes it possible to construct a high-quality WiFi indoor positioning fingerprint database. Firstly, the method utilizes the density clustering method based on Crow Search Algorithm Optimization (CSA-DBSCAN) to process RSS data collected from the reference point. This helps minimize the impact of abnormal RSS data on creating the fingerprint database. Secondly, the RCVAE-GAN depth generation model was developed. The model consists of an encoder E, a generator G, a discriminator D, and a regressor R. After constructing the model, the data with abnormal RSS will be removed and input into the model for pre-training and joint training, resulting in a high-quality deep-generation model. Finally, a high-quality and high-density fingerprint database is constructed by combining the collected reference points with fingerprint data generated by the depth generation model. Experimental results show that the proposed method reduces the root mean square error (RMSE) deviation of the generated fingerprint data by 38% and 12% respectively, compared to the RBF interpolation method and the CVAE-GAN method in the same experimental scenario. The constructed fingerprint database is used for positioning, improving positioning accuracy by 70% and 65% respectively. The method described in this paper can construct a high-quality fingerprint database, effectively improving the efficiency of fingerprint database construction and reducing the costs associated with labor and time.

Conditional Variational Autoencoder for Sign Language Translation with Cross-Modal Alignment

Sign language translation (SLT) aims to convert continuous sign language videos into textual sentences. As a typical multi-modal task, there exists an inherent modality gap between sign language videos and spoken language text, which makes the cross-modal alignment between visual and textual modalities crucial. However, previous studies tend to rely on an intermediate sign gloss representation to help alleviate the cross-modal problem thereby neglecting the alignment across modalities that may lead to compromised results. To address this issue, we propose a novel framework based on Conditional Variational autoencoder for SLT (CV-SLT) that facilitates direct and sufficient cross-modal alignment between sign language videos and spoken language text. Specifically, our CV-SLT consists of two paths with two Kullback-Leibler (KL) divergences to regularize the outputs of the encoder and decoder, respectively. In the prior path, the model solely relies on visual information to predict the target text; whereas in the posterior path, it simultaneously encodes visual information and textual knowledge to reconstruct the target text. The first KL divergence optimizes the conditional variational autoencoder and regularizes the encoder outputs, while the second KL divergence performs a self-distillation from the posterior path to the prior path, ensuring the consistency of decoder outputs.We further enhance the integration of textual information to the posterior path by employing a shared Attention Residual Gaussian Distribution (ARGD), which considers the textual information in the posterior path as a residual component relative to the prior path. Extensive experiments conducted on public datasets demonstrate the effectiveness of our framework, achieving new state-of-the-art results while significantly alleviating the cross-modal representation discrepancy. The code and models are available at https://github.com/rzhao-zhsq/CV-SLT.

SocialCVAE: Predicting Pedestrian Trajectory via Interaction Conditioned Latents

Pedestrian trajectory prediction is the key technology in many applications for providing insights into human behavior and anticipating human future motions. Most existing empirical models are explicitly formulated by observed human behaviors using explicable mathematical terms with deterministic nature, while recent work has focused on developing hybrid models combined with learning-based techniques for powerful expressiveness while maintaining explainability. However, the deterministic nature of the learned steering behaviors from the empirical models limits the models' practical performance. To address this issue, this work proposes the social conditional variational autoencoder (SocialCVAE) for predicting pedestrian trajectories, which employs a CVAE to explore behavioral uncertainty in human motion decisions. SocialCVAE learns socially reasonable motion randomness by utilizing a socially explainable interaction energy map as the CVAE's condition, which illustrates the future occupancy of each pedestrian's local neighborhood area. The energy map is generated using an energy-based interaction model, which anticipates the energy cost (i.e., repulsion intensity) of pedestrians' interactions with neighbors. Experimental results on two public benchmarks including 25 scenes demonstrate that SocialCVAE significantly improves prediction accuracy compared with the state-of-the-art methods, with up to 16.85% improvement in Average Displacement Error (ADE) and 69.18% improvement in Final Displacement Error (FDE). Code is available at: https://github.com/ViviXiang/SocialCVAE.

Abstract 1143: Dissection of melanoma and cutaneous lymphoma spatial molecular architecture by multimodal single-cell and spatial transcriptomics with generative AI

Abstract Advances in single-cell multiomic technologies have revolutionized our understanding of how heterogeneous cell types and cell states shape the tumor microenvironment (TME). However, dissociated single cell profiling lacks spatial context. To resolve the organization, cell-cell communication, and granular structures in the TME, new single-cell spatial transcriptomic (ST) characterizations are needed. Here, we dissected the spatial molecular architecture of the melanoma (30 tumors, 30 patients) and transformed cutaneous T-cell lymphoma (tCTCL, 12 tumors, 6 patients) TMEs by Multiplexed Error-Robust FISH (305 gene MERFISH), with benchmarking scRNAseq data from matched tissues (6 melanoma scFFPEseq, 6 tCTCL scVDJ/RNAseq) for cross-platform validation. We profiled 565,122 cells in melanoma and 92584 cells in tCTCL by MERFISH. We interrogated both datasets using a novel computational framework that includes cell typing by Leiden clustering and marker genes, spatial neighborhood and receptor-ligand (R-L) analyses, and spatial/scRNA imputation via deep learning. In the melanoma dataset, we identified tumor cells, major TME cell types, and rare immune cell types (TCF7+ stem-like T-cells and CD3+TCRα+PAX5+CD79a+ B/T dual-expressor lymphocytes enriched in tertiary lymphoid structures). All cell types were validated in scFFPEseq. As immune R-L interactions are critical therapeutic targets, we developed a novel computation method to quantify spatial R-L interactions by accounting for cell-cell distance, identifying spatially informed R-L pairs that differ from dissociation-based inferences. CNV inference from melanoma scFFPEseq revealed intratumoral heterogeneity (ITH). We therefore adapted the conditional variational autoencoder ENVI to simultaneously incorporate scRNA and MERFISH spatial data into a unified latent embedding. ENVI-ITH successfully learned ITH subgroup information, enabling imputation of phylogenetic lineages on spatial MERFISH. CTCL MERFISH data revealed major TME cell types, yet separation of malignant vs benign T-cells by clustering and marker approaches is hindered by their overlapping gene expression. To overcome this barrier, we deployed ENVI to impute patient-specific malignant and benign TCR clonotypes by training on the matched scVDJ/RNA data, yielding an ENVI-TCR pipeline that can delineate malignant vs benign T-cells in situ. We validated the spatial pattern of malignant T-cells by patient-specific in situ TCR probes. In sum, we have interrogated the melanoma and CTCL TME by MERFISH and presented a novel computational framework for robustly co-embedding scRNA data with imaging-based ST to spatially resolve melanoma ITH and malignant/benign T-cell populations. We believe this approach will be highly impactful for melanoma and CTCL and can be broadly applied to other solid tumors and lymphomas. Citation Format: Zichao Liu, Xiaofei Song, Jiang He, Justin He, Jodi Balasi, Jeffrey H. Chuang, Pei-Ling Chen. Dissection of melanoma and cutaneous lymphoma spatial molecular architecture by multimodal single-cell and spatial transcriptomics with generative AI [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1143.

A Source Identification Problem in Magnetics Solved by Means of Deep Learning Methods

In this study, a deep learning-based approach is used to address inverse problems involving the inversion of a magnetic field and the identification of the relevant source, given the field data within a specific subdomain. Three different techniques are proposed: the first one is characterized by the use of a conditional variational autoencoder (CVAE) and a convolutional neural network (CNN); the second one employs the CVAE (its decoder, more specifically) and a fully connected deep artificial neural network; while the third one (mainly used as a comparison) uses a CNN directly operating on the available data without the use of the CVAE. These methods are applied to the magnetostatic problem outlined in the TEAM 35 benchmark problem, and a comparative analysis between them is conducted.

A Model-Based Reinforcement Learning Method with Conditional Variational Auto-Encoder

Model-based reinforcement learning can effectively improve the sample efficiency of reinforcement learning, but the environment model in this method has errors. The model errors can mislead the policy optimization, leading to suboptimal policy. To improve the generalization ability of the environment model, existing methods often use ensemble models or Bayesian models to build the environment model. However, these methods are computationally intensive and complex to update. Since the generated model can describe the stochastic nature of the environment, this paper proposes a model-based reinforcement learning method based on a conditional variational auto-encoder. In this paper, we use a conditional variational auto-encoder to learn task-related representations and apply the generative model to predict environmental changes. Considering the problem of multi-step error accumulation, model adaptation is utilized to minimize the difference between simulated and real data distributions. Furthermore, the experiments verified that the proposed method can learn task-relevant representations and accelerate policy learning.

STO-CVAE: state transition-oriented conditional variational autoencoder for data augmentation in disability classification

The class imbalance problem occurs when there is an unequal distribution of classes in a dataset and is a significant issue in various artificial intelligence applications. This study focuses on the severe multiclass imbalance problem of human activity recognition in rehabilitation exercises for people with disabilities. To overcome this problem, we present a novel human action-centric augmentation method for human skeleton-based pose estimation. This study proposes the state transition-oriented conditional variational autoencoder (STO-CVAE) to capture action patterns in repeated exercises. The proposed approach generates action samples by capturing temporal information of human skeletons to improve the identification of minority disability classes. We conducted experimental studies with a real-world dataset gathered from rehabilitation exercises and confirmed the superiority and effectiveness of the proposed method. Specifically, all investigated classifiers (i.e., random forest, support vector machine, extreme gradient boosting, light gradient boosting machine, and TabNet) trained with the proposed augmentation method outperformed the models trained without augmentation in terms of the F1-score and accuracy, with F1-score showing the most improvement. Overall, the prediction accuracy of most classes was improved; in particular, the prediction accuracy of the minority classes was greatly improved. Hence, the proposed STO-CVAE can be used to improve the accuracy of disability classification in the field of physical medicine and rehabilitation and to provide suitable personal training and rehabilitation exercise programs.

Robust errant beam prognostics with conditional modeling for particle accelerators

Particle accelerators are complex and comprise thousands of components, with many pieces of equipment running at their peak power. Consequently, they can fault and abort operations for numerous reasons, lowering efficiency and science output. To avoid these faults, we apply anomaly detection techniques to predict unusual behavior and perform preemptive actions to improve the total availability. Supervised machine learning (ML) techniques such as siamese neural network models can outperform the often-used unsupervised or semi-supervised approaches for anomaly detection by leveraging the label information. One of the challenges specific to anomaly detection for particle accelerators is the data’s variability due to accelerator configuration changes within a production run of several months. ML models fail at providing accurate predictions when data changes due to changes in the configuration. To address this challenge, we include the configuration settings into our models and training to improve the results. Beam configurations are used as a conditional input for the model to learn any cross-correlation between the data from different conditions and retain its performance. We employ conditional siamese neural network (CSNN) models and conditional variational auto encoder (CVAE) models to predict errant beam pulses at the spallation neutron source under different system configurations and compare their performance. We demonstrate that CSNNs outperform CVAEs in our application.

An Image Classification Method of Unbalanced Ship Coating Defects Based on DCCVAE-ACWGAN-GP

Affected by the improper operation of the workers, environmental changes during drying and curing or the quality of the paint itself, diverse defects are produced during the process of ship painting. The traditional defect recognition method relies on expert knowledge or experience to detect defects, which is not conducive to ensuring the effectiveness of defect recognition. Therefore, this paper proposes an image generation and recognition model which is suitable for small samples. Based on a deep convolutional neural network (DCNN), the model combines a conditional variational autoencoder (DCCVAE) and auxiliary conditional Wasserstein GAN with gradient penalty (ACWGAN-GP) to gradually expand and generate various coating defect images for solving the overfitting problem due to unbalanced data. The DCNN model is trained based on newly generated image data and original image data so as to build a coating defect image classification model suitable for small samples, which is conducive to improving classification performance. The experimental results showed that our proposed model can achieve up to 92.54% accuracy, an F-score of 88.33%, and a G mean value of 91.93%. Compared with traditional data enhancement methods and classification algorithms, our proposed model can identify various defects in the ship painting process more accurately and consistently, which can provide effective theoretical and technical support for ship painting defect detection and has significant engineering research value and application prospects.

Generating Ultrasonic Foliage Echoes with Variational Autoencoders

Navigation through dense foliage presents a fundamental challenge to autonomous systems, and achieving a performance level similar to echolocating bats could have important applications in areas such as forestry and farming. However, the clutter echoes originating from such environments have been difficult to analyze. To study the problem of sonar‐based navigation in dense foliage in simulation, an artificial generation system for leaf impulse responses (IRs) based on variational auto‐encoders is proposed. The system is to aid the construction of artificial foliage echo environments. A dataset of leaf echoes was collected in an anechoic chamber and convolved with the original signal to estimate the IR of each leaf. A modified version of the conditional variational autoencoder ‐ generative adversarial network (cVAE‐GAN) architecture was trained successfully on this dataset to produce a generative model that was conditional on leaf viewing angles, size, and species. The IRs generated by the model capture quantitative and qualitative similarity to the measured IRs. It surpasses the previous state of the art foliage echo model based on reflecting disks. The model's computational efficiency and its success suggest its potential use for simulating large environments of foliage to study bat biosonar and aid in engineering biomimetic sonar devices.

Rapid Generation of Kilonova Light Curves Using Conditional Variational Autoencoder

The discovery of the optical counterpart, along with the gravitational waves (GWs) from GW170817, of the first binary neutron star merger has opened up a new era for multimessenger astrophysics. Combining the GW data with the optical counterpart, also known as AT 2017gfo and classified as a kilonova, has revealed the nature of compact binary merging systems by extracting enriched information about the total binary mass, the mass ratio, the system geometry, and the equation of state. Even though the detection of kilonovae has brought about a revolution in the domain of multimessenger astronomy, there has been only one kilonova from a GW-detected binary neutron star merger event confirmed so far, and this limits the exact understanding of the origin and propagation of the kilonova. Here, we use a conditional variational autoencoder (CVAE) trained on light-curve data from two kilonova models having different temporal lengths, and consequently, generate kilonova light curves rapidly based on physical parameters of our choice with good accuracy. Once the CVAE is trained, the timescale for light-curve generation is of the order of a few milliseconds, which is a speedup of the generation of light curves by 1000 times as compared to the simulation. The mean squared error between the generated and original light curves is typically 0.015 with a maximum of 0.08 for each set of considered physical parameters, while having a maximum of ≈0.6 error across the whole parameter space. Hence, implementing this technique provides fast and reliably accurate results.

BGaitR-Net: An effective neural model for occlusion reconstruction in gait sequences by exploiting the key pose information

Gait recognition in the presence of occlusion is a challenging problem and the solutions proposed to date either lack robustness or depend on several unrealistic constraints. In this work, we propose a Deep Learning framework to detect and reconstruct the occluded frames in a gait sequence. Initially, occlusion detection is done using a VGG-16 network and for each frame the corresponding pose information is represented as a one-hot encoded vector. This vector is next fused with the corresponding spatial information using a Conditional Variational Autoencoder (CVAE) to obtain an effective embedding. Following this, a Bi-directional Long Short Term Memory (Bi-LSTM) is used to predict the occluded frames using the encoded vector sequence. A decoder next transforms these predicted frames back to the image space. Our proposed reconstruction model termed the Bidirectional Gait Reconstruction Network (BGaitR-Net) is formed by stacking the CVAE, Bi-LSTM, and the decoder. The CASIA-B and OU-ISIR LP datasets are used to prepare extensive gallery sets to train each of the above sub-networks and testing is done using synthetically occluded sequences from the CASIA-B data and real-occluded sequences from the TUM-IITKGP data. A thorough evaluation of our work through Dice Score and GEINet-based recognition accuracy for varying degrees of occlusion highlight the effectiveness of our model in generating frames consistent with the temporal gait pattern. Comparative study with other existing gait recognition techniques (with or without occlusion handling mechanism) and with recent Deep Learning-based video frame prediction methods emphasizes the superiority of BGaitR-Net over the others.