oshri halimi - Academia.edu (original) (raw)

Papers by oshri halimi

Handbook of Numerical Analysis, 2019

Abstract During the last decade the trend in image analysis has been to shift from axiomatically ... more Abstract During the last decade the trend in image analysis has been to shift from axiomatically derived measures into ones that are extracted empirically from data samples. The problem is that accurate geometric data are often unavailable and thus, for proper data augmentation, the research community has resorted yet again to axiomatic construction of invariant measures for curves and surfaces. For some geometric problems, such as shape classification and matching, it is appealing to adopt learning approaches due to their potential accuracy and computational efficiency. Nevertheless, even within the deep learning arena, geometric invariants offer a natural criterion for the learning process. Using geometric invariants one can overcome the need for annotated data and replace it by a purely geometric measure, leading to an unsupervised or a semisupervised learning schemes. Here, we review such constructions that are useful for the geometric analysis of visual information. The measures we explore include the construction of a scale or similarity invariant arc-length for curves and surfaces, an affine invariant one, resulting spectral geometries, and potential signatures that reflect the result of the discrepancies between the corresponding metric spaces. As an example we study novel signatures for surfaces known as the self functional map, which allow us to translate the problem of shape matching into that of computing distances between matrices.

Four corresponding eigenfunctions textured mapped to the same individual at two different poses (... more Four corresponding eigenfunctions textured mapped to the same individual at two different poses (top and bottom), before (pink background) and after (green background) alignment.The first four on the left are eigenfunctions 2-5, followed by 12-15, and 22-25 on the right.

Dense correspondence between articulated objects obtained with the proposed unsupervised loss, op... more Dense correspondence between articulated objects obtained with the proposed unsupervised loss, optimized on a single (unlabeled) example. Our method is compared with the state-of-the-art supervised network pre-trained on human shapes, as well as with two axiomatic methods, employing a post processing algorithm [49] on the axiomatic results. See Section 5.1 for more details.

arXiv (Cornell University), Dec 6, 2018

We introduce the first completely unsupervised correspondence learning approach for deformable 3D... more We introduce the first completely unsupervised correspondence learning approach for deformable 3D shapes. Key to our model is the understanding that natural deformations (such as changes in pose) approximately preserve the metric structure of the surface, yielding a natural criterion to drive the learning process toward distortion-minimizing predictions. On this basis, we overcome the need for annotated data and replace it by a purely geometric criterion. The resulting learning model is class-agnostic, and is able to leverage any type of deformable geometric data for the training phase. In contrast to existing supervised approaches which specialize on the class seen at training time, we demonstrate stronger generalization as well as applicability to a variety of challenging settings. We showcase our method on a wide selection of correspondence benchmarks, where we outperform other methods in terms of accuracy, generalization, and efficiency.

arXiv (Cornell University), Apr 22, 2018

Figure 1: Self functional maps of two classes of articulated objects at different poses. The self... more Figure 1: Self functional maps of two classes of articulated objects at different poses. The self functional map (7 × 7 matrix) of each shape is presented as a gray valued image behind its corresponding object.

arXiv (Cornell University), Jan 27, 2020

According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum of it... more According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum of its parts". This observation was adopted to explain human perception by the Gestalt psychology school of thought in the twentieth century. Here, we claim that observing part of an object which was previously acquired as a whole, one could deal with both partial matching and shape completion in a holistic manner. More specifically, given the geometry of a full, articulated object in a given pose, as well as a partial scan of the same object in a different pose, we address the problem of matching the part to the whole while simultaneously reconstructing the new pose from its partial observation. Our approach is data-driven, and takes the form of a Siamese autoencoder without the requirement of a consistent vertex labeling at inference time; as such, it can be used on unorganized point clouds as well as on triangle meshes. We demonstrate the practical effectiveness of our model in the applications of single-view deformable shape completion and dense shape correspondence, both on synthetic and real-world geometric data, where we outperform prior work on these tasks by a large margin.

ACM Transactions on Graphics, Nov 30, 2022

We propose a novel multi-view camera pipeline for the reconstruction and registration of dynamic ... more We propose a novel multi-view camera pipeline for the reconstruction and registration of dynamic clothing. Our proposed method relies on a specifically designed pattern that allows for precise video tracking in each camera view. We triangulate the tracked points and register the cloth surface in a fine-grained geometric resolution and low localization error. Compared to state-of-the-art methods, our registration exhibits stable correspondence, tracking the same points on the deforming cloth surface along the temporal sequence. As an application, we demonstrate how the use of our registration pipeline greatly improves state-of-the-art pose-based drivable cloth models. Furthermore, we propose a novel model, Garment Avatar, for driving cloth from a dense tracking signal which is obtained from two opposing camera views. The method produces realistic reconstructions which are faithful to the actual geometry of the deforming cloth. In this setting, the user wears a garment with our custom pattern which enables our driving model to reconstruct the geometry. Our code and data are available at https://github.com/HalimiOshri/Pattern-Based-Cloth-Registration-and-Sparse-View-Animation. The released data includes our pattern and registered mesh sequences containing four different subjects and 15k frames in total.

arXiv (Cornell University), Jan 27, 2023

arXiv (Cornell University), Sep 8, 2022

Foundation Models (FMs) have demonstrated unprecedented capabilities including zero-shot learning... more Foundation Models (FMs) have demonstrated unprecedented capabilities including zero-shot learning, high fidelity data synthesis, and out of domain generalization. However, as we show in this paper, FMs still have poor out-of-the-box performance on expert tasks (e.g. retrieval of car manuals technical illustrations from language queries), data for which is either unseen or belonging to a long-tail part of the data distribution of the huge datasets used for FM pre-training. This underlines the necessity to explicitly evaluate and finetune FMs on such expert tasks, arguably ones that appear the most in practical real-world applications. In this paper, we propose a first of its kind FETA benchmark built around the task of teaching FMs to understand technical documentation, via learning to match their graphical illustrations to corresponding language descriptions. Our FETA benchmark focuses on text-to-image and image-to-text retrieval in public car manuals and sales catalogue brochures. FETA is equipped with a procedure for completely automatic annotation extraction (code would be released upon acceptance), allowing easy extension of FETA to more documentation types and application domains in the future. Our automatic annotation leads to an automated performance metric shown to be consistent with metrics computed on human-curated annotations (also released). We provide multiple baselines and analysis of popular FMs on FETA leading to several interesting findings that we believe would be very valuable to the FM community, paving the way towards real-world application of FMs for practical expert tasks currently "overlooked" by standard benchmarks focusing on common objects. 36th Conference on Neural Information Processing Systems (NeurIPS 2022) Track on Datasets and Benchmarks.

arXiv (Cornell University), Dec 6, 2018

We introduce the first completely unsupervised correspondence learning approach for deformable 3D... more We introduce the first completely unsupervised correspondence learning approach for deformable 3D shapes. Key to our model is the understanding that natural deformations (such as changes in pose) approximately preserve the metric structure of the surface, yielding a natural criterion to drive the learning process toward distortion-minimizing predictions. On this basis, we overcome the need for annotated data and replace it by a purely geometric criterion. The resulting learning model is class-agnostic, and is able to leverage any type of deformable geometric data for the training phase. In contrast to existing supervised approaches which specialize on the class seen at training time, we demonstrate stronger generalization as well as applicability to a variety of challenging settings. We showcase our method on a wide selection of correspondence benchmarks, where we outperform other methods in terms of accuracy, generalization, and efficiency.

arXiv (Cornell University), Jun 7, 2022

Computer Vision – ECCV 2020, 2020

In this paper, we advocate the adoption of metric preservation as a powerful prior for learning l... more In this paper, we advocate the adoption of metric preservation as a powerful prior for learning latent representations of deformable 3D shapes. Key to our construction is the introduction of a geometric distortion criterion, defined directly on the decoded shapes, translating the preservation of the metric on the decoding to the formation of linear paths in the underlying latent space. Our rationale lies in the observation that training samples alone are often insufficient to endow generative models with high fidelity, motivating the need for large training datasets. In contrast, metric preservation provides a rigorous way to control the amount of geometric distortion incurring in the construction of the latent space, leading in turn to synthetic samples of higher quality. We further demonstrate, for the first time, the adoption of differentiable intrinsic distances in the backpropagation of a geodesic loss. Our geometric priors are particularly relevant in the presence of scarce training data, where learning any meaningful latent structure can be especially challenging. The effectiveness and potential of our generative model is showcased in applications of style transfer, content generation, and shape completion.

2018 International Conference on 3D Vision (3DV), 2018

Figure 1: Self functional maps of two classes of articulated objects at different poses. The self... more Figure 1: Self functional maps of two classes of articulated objects at different poses. The self functional map (7 × 7 matrix) of each shape is presented as a gray valued image behind its corresponding object.

According to Aristotle, “the whole is greater than the sum of its parts”. This statement was adop... more According to Aristotle, “the whole is greater than the sum of its parts”. This statement was adopted to explain human perception by the Gestalt psychology school of thought in the twentieth century. Here, we claim that when observing a part of an object which was previously acquired as a whole, one could deal with both partial correspondence and shape completion in a holistic manner. More specifically, given the geometry of a full, articulated object in a given pose, as well as a partial scan of the same object in a different pose, we address the new problem of matching the part to the whole while simultaneously reconstructing the new pose from its partial observation. Our approach is data-driven and takes the form of a Siamese autoencoder without the requirement of a consistent vertex labeling at inference time; as such, it can be used on unorganized point clouds as well as on triangle meshes. We demonstrate the practical effectiveness of our model in the applications of single-vie...

Four corresponding eigenfunctions textured mapped to the same individual at two different poses (... more Four corresponding eigenfunctions textured mapped to the same individual at two different poses (top and bottom), before (pink background) and after (green background) alignment.The first four on the left are eigenfunctions 2-5, followed by 12-15, and 22-25 on the right.

Handbook of Numerical Analysis, 2019

During the last decade the trend in image analysis has been to shift from axiomatically derived m... more During the last decade the trend in image analysis has been to shift from axiomatically derived measures into ones that are extracted empirically from data samples. The problem is that accurate geometric data are often unavailable and thus, for proper data augmentation, the research community has resorted yet again to axiomatic construction of invariant measures for curves and surfaces. For some geometric problems, such as shape classification and matching, it is appealing to adopt learning approaches due to their potential accuracy and computational efficiency. Nevertheless, even within the deep learning arena, geometric invariants offer a natural criterion for the learning process. Using geometric invariants one can overcome the need for annotated data and replace it by a purely geometric measure, leading to an unsupervised or a semisupervised learning schemes. Here, we review such constructions that are useful for the geometric analysis of visual information. The measures we expl...

ArXiv, 2020

According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum ... more According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum of its parts". This observation was adopted to explain human perception by the Gestalt psychology school of thought in the twentieth century. Here, we claim that observing part of an object which was previously acquired as a whole, one could deal with both partial matching and shape completion in a holistic manner. More specifically, given the geometry of a full, articulated object in a given pose, as well as a partial scan of the same object in a different pose, we address the problem of matching the part to the whole while simultaneously reconstructing the new pose from its partial observation. Our approach is data-driven, and takes the form of a Siamese autoencoder without the requirement of a consistent vertex labeling at inference time; as such, it can be used on unorganized point clouds as well as on triangle meshes. We demonstrate the practical effectiveness of our model in t...

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019

Dense correspondence between articulated objects obtained with the proposed unsupervised loss, op... more Dense correspondence between articulated objects obtained with the proposed unsupervised loss, optimized on a single (unlabeled) example. Our method is compared with the state-of-the-art supervised network pre-trained on human shapes, as well as with two axiomatic methods, employing a post processing algorithm [49] on the axiomatic results. See Section 5.1 for more details.

Journal of Physics: Conference Series, 2018

Although simulated annealing has become a useful tool for optimization of many systems, its initi... more Although simulated annealing has become a useful tool for optimization of many systems, its initial raison d'etre of achieving the groundstate structure for a spin or atomic/molecular condensed system remains important. Such modelling, whether using analog models such as glass beads or by invoking simple computer models can be suited to undergraduate projects. In this paper we discuss the application of simulated annealing to find the groundstate of a system of liquid crystals (LC) with suspended colloids. These systems are expected to have interesting conductive behaviour, relevant to applications for television and computer screens. In our first stage, a pure LC system was simulated in python and vizualized by undergraduates and presented on an educational website. In the next stage colloid(s) were added, and the original code modified accordingly. Interesting effects such as ordering around the colloid have been seen and will be described. In the final stage and in order to study larger samples, the code was rewritten in C++ and several algorithmic modifications were made. Speed up factors between 100 and more than 1000 were obtained, and fascinating closed cells surrounding the colloids were observed.

Handbook of Numerical Analysis, 2019

Abstract During the last decade the trend in image analysis has been to shift from axiomatically ... more Abstract During the last decade the trend in image analysis has been to shift from axiomatically derived measures into ones that are extracted empirically from data samples. The problem is that accurate geometric data are often unavailable and thus, for proper data augmentation, the research community has resorted yet again to axiomatic construction of invariant measures for curves and surfaces. For some geometric problems, such as shape classification and matching, it is appealing to adopt learning approaches due to their potential accuracy and computational efficiency. Nevertheless, even within the deep learning arena, geometric invariants offer a natural criterion for the learning process. Using geometric invariants one can overcome the need for annotated data and replace it by a purely geometric measure, leading to an unsupervised or a semisupervised learning schemes. Here, we review such constructions that are useful for the geometric analysis of visual information. The measures we explore include the construction of a scale or similarity invariant arc-length for curves and surfaces, an affine invariant one, resulting spectral geometries, and potential signatures that reflect the result of the discrepancies between the corresponding metric spaces. As an example we study novel signatures for surfaces known as the self functional map, which allow us to translate the problem of shape matching into that of computing distances between matrices.

Four corresponding eigenfunctions textured mapped to the same individual at two different poses (... more Four corresponding eigenfunctions textured mapped to the same individual at two different poses (top and bottom), before (pink background) and after (green background) alignment.The first four on the left are eigenfunctions 2-5, followed by 12-15, and 22-25 on the right.

Dense correspondence between articulated objects obtained with the proposed unsupervised loss, op... more Dense correspondence between articulated objects obtained with the proposed unsupervised loss, optimized on a single (unlabeled) example. Our method is compared with the state-of-the-art supervised network pre-trained on human shapes, as well as with two axiomatic methods, employing a post processing algorithm [49] on the axiomatic results. See Section 5.1 for more details.

arXiv (Cornell University), Dec 6, 2018

We introduce the first completely unsupervised correspondence learning approach for deformable 3D... more We introduce the first completely unsupervised correspondence learning approach for deformable 3D shapes. Key to our model is the understanding that natural deformations (such as changes in pose) approximately preserve the metric structure of the surface, yielding a natural criterion to drive the learning process toward distortion-minimizing predictions. On this basis, we overcome the need for annotated data and replace it by a purely geometric criterion. The resulting learning model is class-agnostic, and is able to leverage any type of deformable geometric data for the training phase. In contrast to existing supervised approaches which specialize on the class seen at training time, we demonstrate stronger generalization as well as applicability to a variety of challenging settings. We showcase our method on a wide selection of correspondence benchmarks, where we outperform other methods in terms of accuracy, generalization, and efficiency.

arXiv (Cornell University), Apr 22, 2018

Figure 1: Self functional maps of two classes of articulated objects at different poses. The self... more Figure 1: Self functional maps of two classes of articulated objects at different poses. The self functional map (7 × 7 matrix) of each shape is presented as a gray valued image behind its corresponding object.

arXiv (Cornell University), Jan 27, 2020

According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum of it... more According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum of its parts". This observation was adopted to explain human perception by the Gestalt psychology school of thought in the twentieth century. Here, we claim that observing part of an object which was previously acquired as a whole, one could deal with both partial matching and shape completion in a holistic manner. More specifically, given the geometry of a full, articulated object in a given pose, as well as a partial scan of the same object in a different pose, we address the problem of matching the part to the whole while simultaneously reconstructing the new pose from its partial observation. Our approach is data-driven, and takes the form of a Siamese autoencoder without the requirement of a consistent vertex labeling at inference time; as such, it can be used on unorganized point clouds as well as on triangle meshes. We demonstrate the practical effectiveness of our model in the applications of single-view deformable shape completion and dense shape correspondence, both on synthetic and real-world geometric data, where we outperform prior work on these tasks by a large margin.

ACM Transactions on Graphics, Nov 30, 2022

We propose a novel multi-view camera pipeline for the reconstruction and registration of dynamic ... more We propose a novel multi-view camera pipeline for the reconstruction and registration of dynamic clothing. Our proposed method relies on a specifically designed pattern that allows for precise video tracking in each camera view. We triangulate the tracked points and register the cloth surface in a fine-grained geometric resolution and low localization error. Compared to state-of-the-art methods, our registration exhibits stable correspondence, tracking the same points on the deforming cloth surface along the temporal sequence. As an application, we demonstrate how the use of our registration pipeline greatly improves state-of-the-art pose-based drivable cloth models. Furthermore, we propose a novel model, Garment Avatar, for driving cloth from a dense tracking signal which is obtained from two opposing camera views. The method produces realistic reconstructions which are faithful to the actual geometry of the deforming cloth. In this setting, the user wears a garment with our custom pattern which enables our driving model to reconstruct the geometry. Our code and data are available at https://github.com/HalimiOshri/Pattern-Based-Cloth-Registration-and-Sparse-View-Animation. The released data includes our pattern and registered mesh sequences containing four different subjects and 15k frames in total.

arXiv (Cornell University), Jan 27, 2023

arXiv (Cornell University), Sep 8, 2022

Foundation Models (FMs) have demonstrated unprecedented capabilities including zero-shot learning... more Foundation Models (FMs) have demonstrated unprecedented capabilities including zero-shot learning, high fidelity data synthesis, and out of domain generalization. However, as we show in this paper, FMs still have poor out-of-the-box performance on expert tasks (e.g. retrieval of car manuals technical illustrations from language queries), data for which is either unseen or belonging to a long-tail part of the data distribution of the huge datasets used for FM pre-training. This underlines the necessity to explicitly evaluate and finetune FMs on such expert tasks, arguably ones that appear the most in practical real-world applications. In this paper, we propose a first of its kind FETA benchmark built around the task of teaching FMs to understand technical documentation, via learning to match their graphical illustrations to corresponding language descriptions. Our FETA benchmark focuses on text-to-image and image-to-text retrieval in public car manuals and sales catalogue brochures. FETA is equipped with a procedure for completely automatic annotation extraction (code would be released upon acceptance), allowing easy extension of FETA to more documentation types and application domains in the future. Our automatic annotation leads to an automated performance metric shown to be consistent with metrics computed on human-curated annotations (also released). We provide multiple baselines and analysis of popular FMs on FETA leading to several interesting findings that we believe would be very valuable to the FM community, paving the way towards real-world application of FMs for practical expert tasks currently "overlooked" by standard benchmarks focusing on common objects. 36th Conference on Neural Information Processing Systems (NeurIPS 2022) Track on Datasets and Benchmarks.

arXiv (Cornell University), Dec 6, 2018

We introduce the first completely unsupervised correspondence learning approach for deformable 3D... more We introduce the first completely unsupervised correspondence learning approach for deformable 3D shapes. Key to our model is the understanding that natural deformations (such as changes in pose) approximately preserve the metric structure of the surface, yielding a natural criterion to drive the learning process toward distortion-minimizing predictions. On this basis, we overcome the need for annotated data and replace it by a purely geometric criterion. The resulting learning model is class-agnostic, and is able to leverage any type of deformable geometric data for the training phase. In contrast to existing supervised approaches which specialize on the class seen at training time, we demonstrate stronger generalization as well as applicability to a variety of challenging settings. We showcase our method on a wide selection of correspondence benchmarks, where we outperform other methods in terms of accuracy, generalization, and efficiency.

arXiv (Cornell University), Jun 7, 2022

Computer Vision – ECCV 2020, 2020

In this paper, we advocate the adoption of metric preservation as a powerful prior for learning l... more In this paper, we advocate the adoption of metric preservation as a powerful prior for learning latent representations of deformable 3D shapes. Key to our construction is the introduction of a geometric distortion criterion, defined directly on the decoded shapes, translating the preservation of the metric on the decoding to the formation of linear paths in the underlying latent space. Our rationale lies in the observation that training samples alone are often insufficient to endow generative models with high fidelity, motivating the need for large training datasets. In contrast, metric preservation provides a rigorous way to control the amount of geometric distortion incurring in the construction of the latent space, leading in turn to synthetic samples of higher quality. We further demonstrate, for the first time, the adoption of differentiable intrinsic distances in the backpropagation of a geodesic loss. Our geometric priors are particularly relevant in the presence of scarce training data, where learning any meaningful latent structure can be especially challenging. The effectiveness and potential of our generative model is showcased in applications of style transfer, content generation, and shape completion.

2018 International Conference on 3D Vision (3DV), 2018

Figure 1: Self functional maps of two classes of articulated objects at different poses. The self... more Figure 1: Self functional maps of two classes of articulated objects at different poses. The self functional map (7 × 7 matrix) of each shape is presented as a gray valued image behind its corresponding object.

According to Aristotle, “the whole is greater than the sum of its parts”. This statement was adop... more According to Aristotle, “the whole is greater than the sum of its parts”. This statement was adopted to explain human perception by the Gestalt psychology school of thought in the twentieth century. Here, we claim that when observing a part of an object which was previously acquired as a whole, one could deal with both partial correspondence and shape completion in a holistic manner. More specifically, given the geometry of a full, articulated object in a given pose, as well as a partial scan of the same object in a different pose, we address the new problem of matching the part to the whole while simultaneously reconstructing the new pose from its partial observation. Our approach is data-driven and takes the form of a Siamese autoencoder without the requirement of a consistent vertex labeling at inference time; as such, it can be used on unorganized point clouds as well as on triangle meshes. We demonstrate the practical effectiveness of our model in the applications of single-vie...

Four corresponding eigenfunctions textured mapped to the same individual at two different poses (... more Four corresponding eigenfunctions textured mapped to the same individual at two different poses (top and bottom), before (pink background) and after (green background) alignment.The first four on the left are eigenfunctions 2-5, followed by 12-15, and 22-25 on the right.

Handbook of Numerical Analysis, 2019

During the last decade the trend in image analysis has been to shift from axiomatically derived m... more During the last decade the trend in image analysis has been to shift from axiomatically derived measures into ones that are extracted empirically from data samples. The problem is that accurate geometric data are often unavailable and thus, for proper data augmentation, the research community has resorted yet again to axiomatic construction of invariant measures for curves and surfaces. For some geometric problems, such as shape classification and matching, it is appealing to adopt learning approaches due to their potential accuracy and computational efficiency. Nevertheless, even within the deep learning arena, geometric invariants offer a natural criterion for the learning process. Using geometric invariants one can overcome the need for annotated data and replace it by a purely geometric measure, leading to an unsupervised or a semisupervised learning schemes. Here, we review such constructions that are useful for the geometric analysis of visual information. The measures we expl...

ArXiv, 2020

According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum ... more According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum of its parts". This observation was adopted to explain human perception by the Gestalt psychology school of thought in the twentieth century. Here, we claim that observing part of an object which was previously acquired as a whole, one could deal with both partial matching and shape completion in a holistic manner. More specifically, given the geometry of a full, articulated object in a given pose, as well as a partial scan of the same object in a different pose, we address the problem of matching the part to the whole while simultaneously reconstructing the new pose from its partial observation. Our approach is data-driven, and takes the form of a Siamese autoencoder without the requirement of a consistent vertex labeling at inference time; as such, it can be used on unorganized point clouds as well as on triangle meshes. We demonstrate the practical effectiveness of our model in t...

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019

Dense correspondence between articulated objects obtained with the proposed unsupervised loss, op... more Dense correspondence between articulated objects obtained with the proposed unsupervised loss, optimized on a single (unlabeled) example. Our method is compared with the state-of-the-art supervised network pre-trained on human shapes, as well as with two axiomatic methods, employing a post processing algorithm [49] on the axiomatic results. See Section 5.1 for more details.

Journal of Physics: Conference Series, 2018

Although simulated annealing has become a useful tool for optimization of many systems, its initi... more Although simulated annealing has become a useful tool for optimization of many systems, its initial raison d'etre of achieving the groundstate structure for a spin or atomic/molecular condensed system remains important. Such modelling, whether using analog models such as glass beads or by invoking simple computer models can be suited to undergraduate projects. In this paper we discuss the application of simulated annealing to find the groundstate of a system of liquid crystals (LC) with suspended colloids. These systems are expected to have interesting conductive behaviour, relevant to applications for television and computer screens. In our first stage, a pure LC system was simulated in python and vizualized by undergraduates and presented on an educational website. In the next stage colloid(s) were added, and the original code modified accordingly. Interesting effects such as ordering around the colloid have been seen and will be described. In the final stage and in order to study larger samples, the code was rewritten in C++ and several algorithmic modifications were made. Speed up factors between 100 and more than 1000 were obtained, and fascinating closed cells surrounding the colloids were observed.