Facial Animation Research Articles

Turkey vultures tune their airspeed to changing air density.

Animals must tune their physical performance to changing environmental conditions, and the breadth of environmental tolerance may contribute to delineating the geographic range of a species. A common environmental challenge that flying animals face is the reduction of air density at high elevation and the reduction in the effectiveness of lift production that accompanies it. As a species, turkey vultures (Cathartes aura) inhabit a >3000 m elevation range, and fly considerably higher, necessitating that they accommodate for a 27% change in air density (0.890 to 1.227 kg m-3) through behavior, physiology or biomechanics. We predicted that birds flying at high elevation would maintain aerodynamic lift performance behaviorally via higher flight speeds, rather than increases in power output or local phenotypic adaptation. We used three-dimensional videography to track turkey vultures flying at three elevations, and data supported the hypothesized negative relationship between median airspeed and air density. Additionally, neither the ratio of horizontal speed to sinking speed nor flapping behavior varied with air density.

Artificial Intelligence for Climate Change Biology: From Data Collection to Predictions.

In the era of big data, ecological research is experiencing a transformative shift, yet big-data advancements in thermal ecology and the study of animal responses to climate conditions remain limited. This review discusses how big data analytics and artificial intelligence (AI) can significantly enhance our understanding of microclimates and animal behaviors under changing climatic conditions. We explore AI's potential to refine microclimate models and analyze data from advanced sensors and camera technologies, which capture detailed, high-resolution information. This integration can allow researchers to dissect complex ecological and physiological processes with unprecedented precision. We describe how AI can enhance microclimate modeling through improved bias correction and downscaling techniques, providing more accurate estimates of the conditions that animals face under various climate scenarios. Additionally, we explore AI's capabilities in tracking animal responses to these conditions, particularly through innovative classification models that utilize sensors such as accelerometers and acoustic loggers. For example, the widespread usage of camera traps can benefit from AI-driven image classification models to accurately identify thermoregulatory responses, such as shade usage and panting. AI is therefore instrumental in monitoring how animals interact with their environments, offering vital insights into their adaptive behaviors. Finally, we discuss how these advanced data-driven approaches can inform and enhance conservation strategies. In particular, detailed mapping of microhabitats essential for species survival under adverse conditions can guide the design of climate-resilient conservation and restoration programs that prioritize habitat features crucial for biodiversity resilience. In conclusion, the convergence of AI, big data, and ecological science heralds a new era of precision conservation, essential for addressing the global environmental challenges of the 21st century.

A randomised controlled test in virtual reality of the effects on paranoid thoughts of virtual humans’ facial animation and expression

Virtual reality (VR) is increasingly used in the study and treatment of paranoia. This is based on the finding that people who mistakenly perceive hostile intent from other people also perceive similar threat from virtual characters. However, there has been no study of the programming characteristics of virtual characters that may influence their interpretation. We set out to investigate how the animation and expressions of virtual humans may affect paranoia. In a two-by-two factor, between-groups, randomized design, 122 individuals with elevated paranoia rated their perceptions of virtual humans, set in an eye-tracking enabled VR lift scenario, that varied in facial animation (static or animated) and expression (neutral or positive). Both facial animation (group difference = 102.328 [51.783, 152.872], p < 0.001, ηp2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\\eta }_{p}^{2}\\hspace{0.17em}$$\\end{document}= 0.125) and positive expressions (group difference = 53.016 [0.054, 105.979], p = 0.049, ηp2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\\eta }_{p}^{2}\\hspace{0.17em}$$\\end{document}= 0.033) led to less triggering of paranoid thoughts about the virtual humans. Facial animation (group difference = 2.442 [− 4.161, − 0.724], p = 0.006, ηp2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\\eta }_{p}^{2}\\hspace{0.17em}$$\\end{document}= 0.063) but not positive expressions (group difference = 0.344 [− 1.429, 2.110], p = 0.681, ηp2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\\eta }_{p}^{2}\\hspace{0.17em}$$\\end{document}= 0.001) significantly increased the likelihood of neutral thoughts about the characters. Our study shows that the detailed programming of virtual humans can impact the occurrence of paranoid thoughts in VR. The programming of virtual humans needs careful consideration depending on the purpose of their use.

Universal Facial Encoding of Codec Avatars from VR Headsets

Faithful real-time facial animation is essential for avatar-mediated telepresence in Virtual Reality (VR). To emulate authentic communication, avatar animation needs to be efficient and accurate: able to capture both extreme and subtle expressions within a few milliseconds to sustain the rhythm of natural conversations. The oblique and incomplete views of the face, variability in the donning of headsets, and illumination variation due to the environment are some of the unique challenges in generalization to unseen faces. In this paper, we present a method that can animate a photorealistic avatar in realtime from head-mounted cameras (HMCs) on a consumer VR headset. We present a self-supervised learning approach, based on a cross-view reconstruction objective, that enables generalization to unseen users. We present a lightweight expression calibration mechanism that increases accuracy with minimal additional cost to run-time efficiency. We present an improved parameterization for precise ground-truth generation that provides robustness to environmental variation. The resulting system produces accurate facial animation for unseen users wearing VR headsets in realtime. We compare our approach to prior face-encoding methods demonstrating significant improvements in both quantitative metrics and qualitative results.

Learning a Generalized Physical Face Model From Data

Physically-based simulation is a powerful approach for 3D facial animation as the resulting deformations are governed by physical constraints, allowing to easily resolve self-collisions, respond to external forces and perform realistic anatomy edits. Today's methods are data-driven, where the actuations for finite elements are inferred from captured skin geometry. Unfortunately, these approaches have not been widely adopted due to the complexity of initializing the material space and learning the deformation model for each character separately, which often requires a skilled artist followed by lengthy network training. In this work, we aim to make physics-based facial animation more accessible by proposing a generalized physical face model that we learn from a large 3D face dataset. Once trained, our model can be quickly fit to any unseen identity and produce a ready-to-animate physical face model automatically. Fitting is as easy as providing a single 3D face scan, or even a single face image. After fitting, we offer intuitive animation controls, as well as the ability to retarget animations across characters. All the while, the resulting animations allow for physical effects like collision avoidance, gravity, paralysis, bone reshaping and more.

Formulating facial mesh tracking as a differentiable optimization problem: a backpropagation-based solution

Facial mesh tracking enables the production of topologically consistent 3D facial meshes from stereo video input captured by calibrated cameras. This technology is an integral part of many digital human applications, such as personalized avatar creation, audio-driven 3D facial animation, and talking face video generation. Currently, most facial mesh tracking methods are built on computer graphics techniques, which involve complex procedures and often necessitate human annotation within pipelines. As a result, these approaches are difficult to implement and hard to generalize across various scenarios. We propose a backpropagation-based solution that formulates facial mesh tracking as a differentiable optimization problem called the BPMT. Our solution leverages visual clues extracted from the stereo input to estimate vertex-wise geometry and texture information. The BPMT is composed of two steps: automatic face analysis and mesh tracking. In the first step, a range of visual clues are automatically extracted from the input, including facial point clouds, multi-view 2D landmarks, 3D landmarks in the world coordinate system, motion fields, and image masks. The second step can be viewed as a differentiable optimization problem, with constraints comprising stereo video input and facial clues. The primary objective is to achieve topologically consistent 3D facial meshes across frames. Additionally, the parameters to be optimized encompass the positions of free-form deformed vertices and a shared texture UV map. Furthermore, the 3D morphable model (3DMM) is introduced as a form of regularization to enhance the convergence of the optimization process. Leveraging the fully developed backpropagation software, we progressively register the facial meshes to the recorded object, generating high-quality 3D faces with consistent topologies. The BPMT requires no manual labeling within the pipeline, making it suitable for producing large-scale stereo facial data. Moreover, our method exhibits a high degree of flexibility and extensibility, positioning it as a promising platform for future research in the community.

DiffPoseTalk: Speech-Driven Stylistic 3D Facial Animation and Head Pose Generation via Diffusion Models

The generation of stylistic 3D facial animations driven by speech presents a significant challenge as it requires learning a many-to-many mapping between speech, style, and the corresponding natural facial motion. However, existing methods either employ a deterministic model for speech-to-motion mapping or encode the style using a one-hot encoding scheme. Notably, the one-hot encoding approach fails to capture the complexity of the style and thus limits generalization ability. In this paper, we propose DiffPoseTalk, a generative framework based on the diffusion model combined with a style encoder that extracts style embeddings from short reference videos. During inference, we employ classifier-free guidance to guide the generation process based on the speech and style. In particular, our style includes the generation of head poses, thereby enhancing user perception. Additionally, we address the shortage of scanned 3D talking face data by training our model on reconstructed 3DMM parameters from a high-quality, in-the-wild audio-visual dataset. Extensive experiments and user study demonstrate that our approach outperforms state-of-the-art methods. The code and dataset are at https://diffposetalk.github.io.

Modulation of the attentional response to baby schema by testosterone in pedohebephilic men and its relation to the nurturing system

Previous fMRI research found increased brain responses in men with pedophilic interest to non-sexual pictures of child and animal faces. This raised the question of whether an aberrant nurturing system could be linked to pedophilia. To further explore this hypothesis, 20 pedohebephilic and 23 teleiophilic men performed a target detection task with adult versus infant human and animal faces, which measured selective attention towards the baby schema by comparing reaction times to infant versus adult targets that were presented amongst distractors of the other category. Since the response to baby schema can be influenced by steroid hormones, saliva samples were additionally collected to determine endogenous testosterone, progesterone, estradiol and cortisol. Contrary to expectations, all men did not react faster to infant than adult faces. Yet, pedohebephilic men were more distracted by infant’s faces than teleiophilic men. Pedohebephilic men with higher testosterone were faster in orienting attention to infant targets in the context of adult distractors. This association was not observed in teleiophilic men. Our results support the idea of an overactive nurturing system in pedophilia, which may be influenced by the endogenous testosterone level.

Near-realtime Facial Animation by Deep 3D Simulation Super-Resolution

Near-realtime Facial Animation by Deep 3D Simulation Super-Resolution

Facial Animation Strategies for Improved Emotional Expression in Virtual Reality

The portrayal of emotions by virtual characters is crucial in virtual reality (VR) communication. Effective communication in VR relies on a shared understanding, which is significantly enhanced when virtual characters authentically express emotions that align with their spoken words. While human emotions are often conveyed through facial expressions, existing facial animation techniques have mainly focused on lip-syncing and head movements to improve naturalness. This study investigates the influence of various factors in facial animation on the emotional representation of virtual characters. We conduct a comparative and analytical study using an audio-visual database, examining the impact of different animation factors. To this end, we utilize a total of 24 voice samples, representing 12 different speakers, with each emotional voice segment lasting approximately 4–5 s. Using these samples, we design six perceptual experiments to investigate the impact of facial cues—including facial expression, lip movement, head motion, and overall appearance—on the expression of emotions by virtual characters. Additionally, we engaged 20 participants to evaluate and select appropriate combinations of facial expressions, lip movements, head motions, and appearances that align with the given emotion and its intensity. Our findings indicate that emotional representation in virtual characters is closely linked to facial expressions, head movements, and overall appearance. Conversely, lip-syncing, which has been a primary focus in prior studies, seems less critical for conveying emotions, as its accuracy is difficult to perceive with the naked eye. The results of our study can significantly benefit the VR community by aiding in the development of virtual characters capable of expressing a diverse range of emotions.

AnaConDaR: Anatomically-Constrained Data-Adaptive Facial Retargeting

Offline facial retargeting, i.e., transferring facial expressions from a source to a target character, is a common production task that still regularly leads to considerable algorithmic challenges. This task can be roughly dissected into the transfer of sequential facial animations and non-sequential blendshape personalization. Both problems are typically solved by data-driven methods that require an extensive corpus of costly target examples. Other than that, geometrically motivated approaches do not require intensive data collection but cannot account for character-specific deformations and are known to cause manifold visual artifacts.We present AnaConDaR, a novel method for offline facial retargeting, as a hybrid of data-driven and geometry-driven methods that incorporates anatomical constraints through a physics-based simulation. As a result, our approach combines the advantages of both paradigms while balancing out the respective disadvantages. In contrast to other recent concepts, AnaConDaR achieves substantially individualized results even when only a handful of target examples are available. At the same time, we do not make the common assumption that for each target example a matching source expression must be known. Instead, AnaConDaR establishes correspondences between the source and the target character by a data-driven embedding of the target examples in the source domain. We evaluate our offline facial retargeting algorithm visually, quantitatively, and in two user studies.

What enlightenment has the development of lung cancer bone metastasis brought in the last 22 years.

Lung cancer bone metastasis (LCBM) is a disease with a poor prognosis, high risk and large patient population. Although considerable scientific output has accumulated on LCBM, problems have emerged, such as confusing research structures. To organize the research frontiers and body of knowledge of the studies on LCBM from the last 22 years according to their basic research and translation, clinical treatment, and clinical diagnosis to provide a reference for the development of new LCBM clinical and basic research. We used tools, including R, VOSviewer and CiteSpace software, to measure and visualize the keywords and other metrics of 1903 articles from the Web of Science Core Collection. We also performed enrichment and protein-protein interaction analyses of gene expression datasets from LCBM cases worldwide. Research on LCBM has received extensive attention from scholars worldwide over the last 20 years. Targeted therapies and immunotherapies have evolved into the mainstream basic and clinical research directions. The basic aspects of drug resistance mechanisms and parathyroid hormone-related protein may provide new ideas for mechanistic study and improvements in LCBM prognosis. The produced molecular map showed that ribosomes and focal adhesion are possible pathways that promote LCBM occurrence. Novel therapies for LCBM face animal testing and drug resistance issues. Future focus should centre on advancing clinical therapies and researching drug resistance mechanisms and ribosome-related pathways.

Bigger isn't always better: Challenging assumptions about the associations between diapause, body weight, and overwintering survival.

During the winter, animals face limited food availability. Many animals enter dormancy to reduce their winter energy expenditure. Most insects spend the winter in diapause, a state of programmed dormancy. It is often assumed that diapausing insects need nutrient stores to fuel their many months of basal metabolism and must grow heavier than their non-diapause-programmed counterparts. However, the extent to which food limitation affects body weight during overwintering preparation as well as the likelihood and duration of diapause remainsunclear. We limited the duration of the feeding period and thus the total quantity of food available to diapause-destined larvae of the pupal-diapausing flesh fly, Sarcophaga crassipalpis, to test how food limitation affects body weight in the context of diapause programming. We also tested the extent to which food deprivation and body weight affect the likelihood and duration of diapause. We hypothesized that diapause-destined larvae grow more quickly and pupariate at a heavier body weight than non-diapause larvae. We also hypothesized that body weight is more dramatically reduced by food limitations when a larva is programmed for diapause. Finally, we hypothesized that larvae with lighter body weight (i.e., food limited) are less likely to enter pupal diapause and also stay in diapause for a shorter duration than heavier, well-fed, individuals. Contrary to our hypotheses that diapausing insects are heavier than their non-diapausing counterparts, we found diapausing pupae weighed less than non-diapausing pupae, especially when larvae received limited food. We found light pupae did not abort their diapause program. In both diapausing and non-diapausing pupae, body weight was positively correlated with simulated winter survival. However, above a weight threshold, body weight no longer affected simulated winter survival in diapausing pupae. Contrary to our predictions and the general consensus in much of the diapause literature, we also found that lighter pupae stayed in diapause longer than heavier pupae. Overall, our results challenge the precept that body weight and diapause are positively associated. The relationship between body weight and diapause is complex and may be affected by the availability of food before and after winter, the availability of high-quality overwintering sites, and the life history of a particular insect.

Svara Rachana - Audio Driven Facial Expression Synthesis

Abstract: Svara Rachana is a fusion of artificial intelligence and facial animation which aims to revolutionize the field of digital communication. Harnessing the ever-evolving power of neural networks in the form of Long Short-Term Memory (LSTM) model, Svara Rachana offers a cutting edge, interactive web application designed to synchronize human speech with realistic 3D facial animation. Users can upload or record an audio file and upload it to the web interface containing human speech, with the core functionality being the generation of synchronized lip movements on a 3D avatar. The system gives special emphasis on the accuracy of the system to generate reliable facial animation movements. By providing an interactive, human like 3D model, Svara Rachana aims to make machine to human interaction a more impactful experience by blurring the lines between humans and machines.

3D facial animation driven by speech-video dual-modal signals

In recent years, the applications of digital humans have become increasingly widespread. One of the most challenging core technologies is the generation of highly realistic and automated 3D facial animation that combines facial movements and speech. The single-modal 3D facial animation driven by speech typically ignores the weak correlation between speech and upper facial movements as well as head posture. In contrast, the video-driven approach can perfectly solve the posture problem while obtaining natural expressions. However, mapping 2D facial information to 3D facial information may lead to information loss, which make lip synchronization generated by video-driven methods is not as good as the speech-driven methods trained on 4D facial data. Therefore, this paper proposes a dual-modal generation method that uses speech and video information to generate more natural and vivid 3D facial animation. Specifically, the lip movements related to speech are generated by speech-video information, while speech-uncorrelated postures and expressions are generated solely by video information. The speech-driven module is used to extract speech features, and its output lip animation is then used as the foundation for facial animation. The expression and pose module is used to extract temporal visual features for regressing expression and head posture parameters. We fuse speech and video features to obtain chin posture parameters related to lip movements, and use these parameters to fine-tune the lip animation generated form the speech-driven module. This paper introduces multiple consistency losses to enhance the network’s capability to generate expressions and postures. Experiments conducted on the LRS3, TCD-TIMIT and MEAD datasets show that the proposed method achieves better performance on evaluation metrics such as CER, WER, VER and VWER than the current state-of-the-art methods. In addition, a perceptual user study show that over 77% and 70% of cases believe that this paper’s method is better than the comparative algorithms EMOCA and SPECTRE in terms of realism. In terms of lip synchronization, it received over 79% and 66% of cases support, respectively. Both evaluation methods demonstrate the effectiveness of the proposed method.

Exploring the Influence of Virtual Avatar Heads in Mixed Reality on Social Presence, Performance and User Experience in Collaborative Tasks.

In Mixed Reality (MR), users' heads are largely (if not completely) occluded by the MR Head-Mounted Display (HMD) they are wearing. As a consequence, one cannot see their facial expressions and other communication cues when interacting locally. In this paper, we investigate how displaying virtual avatars' heads on-top of the (HMD-occluded) heads of participants in a Video See-Through (VST) Mixed Reality local collaborative task could improve their collaboration as well as social presence. We hypothesized that virtual heads would convey more communicative cues (such as eye direction or facial expressions) hidden by the MR HMDs and lead to better collaboration and social presence. To do so, we conducted a between-subject study ($\mathrm{n}=88$) with two independent variables: the type of avatar (CartoonAvatar/RealisticAvatar/NoAvatar) and the level of facial expressions provided (HighExpr/LowExpr). The experiment involved two dyadic communication tasks: (i) the "20-question" game where one participant asks questions to guess a hidden word known by the other participant and (ii) a urban planning problem where participants have to solve a puzzle by collaborating. Each pair of participants performed both tasks using a specific type of avatar and facial animation. Our results indicate that while adding an avatar's head does not necessarily improve social presence, the amount of facial expressions provided through the social interaction does have an impact. Moreover, participants rated their performance higher when observing a realistic avatar but rated the cartoon avatars as less uncanny. Taken together, our results contribute to a better understanding of the role of partial avatars in local MR collaboration and pave the way for further research exploring collaboration in different scenarios, with different avatar types or MR setups.

Facial action units detection using temporal context and feature reassignment

AbstractFacial action units (AUs) encode the activations of facial muscle groups, playing a crucial role in expression analysis and facial animation. However, current deep learning AU detection methods primarily focus on single‐image analysis, which limits the exploitation of rich temporal context for robust outcomes. Moreover, the scale of available datasets remains limited, leading models trained on these datasets to tend to suffer from overfitting issues. This paper proposes a novel AU detection method integrating spatial and temporal data with inter‐subject feature reassignment for accurate and robust AU predictions. Our method first extracts regional features from facial images. Then, to effectively capture both the temporal context and identity‐independent features, we introduce a temporal feature combination and feature reassignment (TC&amp;FR) module, which transforms single‐image features into a cohesive temporal sequence and fuses features across multiple subjects. This transformation encourages the model to utilize identity‐independent features and temporal context, thus ensuring robust prediction outcomes. Experimental results demonstrate the enhancements brought by the proposed modules and the state‐of‐the‐art (SOTA) results achieved by our method.

FACIAL ANIMATION AND EMPATHY IN THE&lt;i&gt; LAST OF US PARTS I AND II&lt;/i&gt;

For more than a decade the videogame studio Naughty Dog has deployed increasingly sophisticated facial animations, greatly expanding its characters’ abilities to convey realistic and compelling emotion. In a parallel effort, the studio has remediated cinematic forms like the closeup, integrating them with the unique affordances of videoludic media. Naughty Dog’s 2020 The Last of Us Part II takes this a step further, making characters’ faces a vital aspect of the game’s interface: the dynamically changing emotional expression of 25 in-game characters offers fine-grained feedback regarding player choices and actions, and it also encourages players to engage with the characters less like tools or targets and more like autonomous human beings. Through a close study of a single game character, Ellie Williams, my article illuminates the narrative and gameplay impact of this merger of face and interface. Ellie began The Last of Us Part I (2013) as a non-player character (NPC) and, in some ways, an archetypal “damsel in distress,” but she evolved to become a fan favourite as well as the sequel’s protagonist and principal player-character (PC). Along the way, Ellie also became something of a feminist icon: she is a queer young woman who wears practical clothing, a character very different than the stereotypical heroes that dominated previous videogame generations. The success of the games and their central character, I argue, turns on their ability to encourage emotional connections, sometimes called empathy, inviting players to engage with videogame characters in ways that parallel their responses to characters in older media forms.

A Qualitative Study: Issue(s) of Stray Animal(s) in Turkey from the Point of View of Volunteers

The aim of this study is to reveal the issue of stray animals from the volunteers’ viewpoint and determine the views and suggestions of volunteers who struggle for the rights of stray animals in Turkey. For this, a study based on qualitative research techniques was carried out with participants who voluntarily work for stray animals. Among the qualitative research patterns, case study pattern, semi-structured interview as data gathering tool, among nonprobable choice method, snowball method as data gathering tool were chosen. All the data which were acquired with semi-structured profoundly interviews were analyzed by descriptive analyzing method. This study does not reveal that fact issue of stray animals in cities of Turkey but reveals the issues of stray animals themselves and also the volunteers who have been striving for stray animals voluntarily. The issue of stray animals in Turkey is rooted mainly in local governments’ executions. It is emphasized one more time that the current executions of local governments are inadequate to be able to solve the issues of stray animals. The most important cause for this is that local governments’ not having applied the process of sterilization effectively and thus, the population of stray animals cannot be taken under control. The other significant causes are inefficiency of animal shelters run by local governments during the treatment, the unawareness of the people in the public and the nonlegal applications done by the local governments through stray animals. According to the volunteers, this issue can be handled by the state. The solution for this issue is effective sterilization and the rising awareness of the people. Eventually, it does not matter whether the solution for the issues about stray animals or the solution for issues which stray animals face, local and central governments have plenty of duties. But the main actor for this is the central government.

3D Face Reconstruction with Feature Enhancement using Bi-FPN for Forensic Analysis

The representation of facial features in three-dimensional space plays a pivotal role in various applications such as facial recognition, virtual reality, and digital entertainment. However, achieving high-fidelity reconstructions from two-dimensional facial images remains a challenging task, particularly in preserving fine texture details. This research addresses this problem by proposing a novel approach that leverages a combination of advanced techniques, including Resnet, Flame model, Bi-FPN, and a differential render architecture. The primary objective of this study is to enhance texture details in reconstructed 3D facial images. The integration of Bi-FPN (Bi-directional Feature Pyramid Network) enhances feature extraction and fusion across multiple scales, facilitating the preservation of texture details across different regions of the face. The objective is to accurately represent facial features from 2D images in three-dimensional space. By combining these methods, the proposed framework achieves significant improvements in preserving fine texture details and overall facial structure. Experimental results demonstrate the effectiveness of the approach, suggesting its potential for various applications such as virtual try-on and facial animation.