Ahmed Elgammal

Kunpeng Song, Ligong Han, Bingchen Liu et al.

Can a text-to-image diffusion model be used as a training objective for adapting a GAN generator to another domain? In this paper, we show that the classifier-free guidance can be leveraged as a critic and enable generators to distill knowledge from large-scale text-to-image diffusion models. Generators can be efficiently shifted into new domains indicated by text prompts without access to groundtruth samples from target domains. We demonstrate the effectiveness and controllability of our method through extensive experiments. Although not trained to minimize CLIP loss, our model achieves equally high CLIP scores and significantly lower FID than prior work on short prompts, and outperforms the baseline qualitatively and quantitatively on long and complicated prompts. To our best knowledge, the proposed method is the first attempt at incorporating large-scale pre-trained diffusion models and distillation sampling for text-driven image generator domain adaptation and gives a quality previously beyond possible. Moreover, we extend our work to 3D-aware style-based generators and DreamBooth guidance.

Kunpeng Song, Yizhe Zhu, Bingchen Liu et al.

In this paper, we present MoMA: an open-vocabulary, training-free personalized image model that boasts flexible zero-shot capabilities. As foundational text-to-image models rapidly evolve, the demand for robust image-to-image translation grows. Addressing this need, MoMA specializes in subject-driven personalized image generation. Utilizing an open-source, Multimodal Large Language Model (MLLM), we train MoMA to serve a dual role as both a feature extractor and a generator. This approach effectively synergizes reference image and text prompt information to produce valuable image features, facilitating an image diffusion model. To better leverage the generated features, we further introduce a novel self-attention shortcut method that efficiently transfers image features to an image diffusion model, improving the resemblance of the target object in generated images. Remarkably, as a tuning-free plug-and-play module, our model requires only a single reference image and outperforms existing methods in generating images with high detail fidelity, enhanced identity-preservation and prompt faithfulness. Our work is open-source, thereby providing universal access to these advancements.

Faizan Farooq Khan, Diana Kim, Divyansh Jha et al.

Discovering the creative potentials of a random signal to various artistic expressions in aesthetic and conceptual richness is a ground for the recent success of generative machine learning as a way of art creation. To understand the new artistic medium better, we conduct a comprehensive analysis to position AI-generated art within the context of human art heritage. Our comparative analysis is based on an extensive dataset, dubbed ``ArtConstellation,'' consisting of annotations about art principles, likability, and emotions for 6,000 WikiArt and 3,200 AI-generated artworks. After training various state-of-the-art generative models, art samples are produced and compared with WikiArt data on the last hidden layer of a deep-CNN trained for style classification. We actively examined the various art principles to interpret the neural representations and used them to drive the comparative knowledge about human and AI-generated art. A key finding in the semantic analysis is that AI-generated artworks are visually related to the principle concepts for modern period art made in 1800-2000. In addition, through Out-Of-Distribution (OOD) and In-Distribution (ID) detection in CLIP space, we find that AI-generated artworks are ID to human art when they depict landscapes and geometric abstract figures, while detected as OOD when the machine art consists of deformed and twisted figures. We observe that machine-generated art is uniquely characterized by incomplete and reduced figuration. Lastly, we conducted a human survey about emotional experience. Color composition and familiar subjects are the key factors of likability and emotions in art appreciation. We propose our whole methodologies and collected dataset as our analytical framework to contrast human and AI-generated art, which we refer to as ``ArtNeuralConstellation''. Code is available at: https://github.com/faixan-khan/ArtNeuralConstellation

Kunpeng Song, Tingbo Hou, Zecheng He et al.

In this paper, we introduce DirectorLLM, a novel video generation model that employs a large language model (LLM) to orchestrate human poses within videos. As foundational text-to-video models rapidly evolve, the demand for high-quality human motion and interaction grows. To address this need and enhance the authenticity of human motions, we extend the LLM from a text generator to a video director and human motion simulator. Utilizing open-source resources from Llama 3, we train the DirectorLLM to generate detailed instructional signals, such as human poses, to guide video generation. This approach offloads the simulation of human motion from the video generator to the LLM, effectively creating informative outlines for human-centric scenes. These signals are used as conditions by the video renderer, facilitating more realistic and prompt-following video generation. As an independent LLM module, it can be applied to different video renderers, including UNet and DiT, with minimal effort. Experiments on automatic evaluation benchmarks and human evaluations show that our model outperforms existing ones in generating videos with higher human motion fidelity, improved prompt faithfulness, and enhanced rendered subject naturalness.

Kunpeng Song, Ahmed Elgammal

Creative image generation has emerged as a compelling area of research, driven by the need to produce novel and high-quality images that expand the boundaries of imagination. In this work, we propose a novel framework for creative generation using diffusion models, where creativity is associated with the inverse probability of an image's existence in the CLIP embedding space. Unlike prior approaches that rely on a manual blending of concepts or exclusion of subcategories, our method calculates the probability distribution of generated images and drives it towards low-probability regions to produce rare, imaginative, and visually captivating outputs. We also introduce pullback mechanisms, achieving high creativity without sacrificing visual fidelity. Extensive experiments on text-to-image diffusion models demonstrate the effectiveness and efficiency of our creative generation framework, showcasing its ability to produce unique, novel, and thought-provoking images. This work provides a new perspective on creativity in generative models, offering a principled method to foster innovation in visual content synthesis.

Bingchen Liu, Yizhe Zhu, Kunpeng Song et al.

Training Generative Adversarial Networks (GAN) on high-fidelity images usually requires large-scale GPU-clusters and a vast number of training images. In this paper, we study the few-shot image synthesis task for GAN with minimum computing cost. We propose a light-weight GAN structure that gains superior quality on 1024*1024 resolution. Notably, the model converges from scratch with just a few hours of training on a single RTX-2080 GPU, and has a consistent performance, even with less than 100 training samples. Two technique designs constitute our work, a skip-layer channel-wise excitation module and a self-supervised discriminator trained as a feature-encoder. With thirteen datasets covering a wide variety of image domains (The datasets and code are available at: https://github.com/odegeasslbc/FastGAN-pytorch), we show our model's superior performance compared to the state-of-the-art StyleGAN2, when data and computing budget are limited.

Bingchen Liu, Yizhe Zhu, Kunpeng Song et al.

Imagining a colored realistic image from an arbitrarily drawn sketch is one of the human capabilities that we eager machines to mimic. Unlike previous methods that either requires the sketch-image pairs or utilize low-quantity detected edges as sketches, we study the exemplar-based sketch-to-image (s2i) synthesis task in a self-supervised learning manner, eliminating the necessity of the paired sketch data. To this end, we first propose an unsupervised method to efficiently synthesize line-sketches for general RGB-only datasets. With the synthetic paired-data, we then present a self-supervised Auto-Encoder (AE) to decouple the content/style features from sketches and RGB-images, and synthesize images that are both content-faithful to the sketches and style-consistent to the RGB-images. While prior works employ either the cycle-consistence loss or dedicated attentional modules to enforce the content/style fidelity, we show AE's superior performance with pure self-supervisions. To further improve the synthesis quality in high resolution, we also leverage an adversarial network to refine the details of synthetic images. Extensive experiments on 1024*1024 resolution demonstrate a new state-of-art-art performance of the proposed model on CelebA-HQ and Wiki-Art datasets. Moreover, with the proposed sketch generator, the model shows a promising performance on style mixing and style transfer, which require synthesized images to be both style-consistent and semantically meaningful. Our code is available on https://github.com/odegeasslbc/Self-Supervised-Sketch-to-Image-Synthesis-PyTorch, and please visit https://create.playform.io/my-projects?mode=sketch for an online demo of our model.

Toufiq Parag, Ahmed Elgammal

The long-context capability of recent large transformer models can be surmised to rely on techniques such as attention/model parallelism, as well as hardware-level optimizations. While these strategies allow input lengths to scale to millions of tokens, they do not fundamentally mitigate the quadratic computational and memory complexity of the core attention mechanism. In this paper, we address the challenge of long-context processing on a single device using generic implementations by reducing the quadratic memory footprint and inference cost. Existing approaches to extend the context length for generic single device implementations -- such as token merging and modified attentions -- are often modality specific and attain a suboptimal tradeoff between accuracy and efficiency. To overcome these limitations, we propose MAELRE (Modality Agnostic Efficient Long Range Encoder), a unified and efficient transformer architecture designed for long-range encoding across diverse modalities. MAELRE integrates token merging with attention approximation, progressively merging tokens at different stages of internal computational blocks. It employs a lightweight attention approximation when the number of tokens is large, and switches to standard dot-product attention as the sequence becomes shorter through successive aggregation. We demonstrate that MAELRE achieves superior accuracy while reducing computational cost compared to existing long-context models on classification tasks spanning multiple modalities, including text, time series, audio, and vision.

Diana Kim, Ahmed Elgammal, Marian Mazzone

We present a machine learning system that can quantify fine art paintings with a set of visual elements and principles of art. This formal analysis is fundamental for understanding art, but developing such a system is challenging. Paintings have high visual complexities, but it is also difficult to collect enough training data with direct labels. To resolve these practical limitations, we introduce a novel mechanism, called proxy learning, which learns visual concepts in paintings though their general relation to styles. This framework does not require any visual annotation, but only uses style labels and a general relationship between visual concepts and style. In this paper, we propose a novel proxy model and reformulate four pre-existing methods in the context of proxy learning. Through quantitative and qualitative comparison, we evaluate these methods and compare their effectiveness in quantifying the artistic visual concepts, where the general relationship is estimated by language models; GloVe or BERT. The language modeling is a practical and scalable solution requiring no labeling, but it is inevitably imperfect. We demonstrate how the new proxy model is robust to the imperfection, while the other models are sensitively affected by it.

Mahyar Khayatkhoei, Ahmed Elgammal

As the success of Generative Adversarial Networks (GANs) on natural images quickly propels them into various real-life applications across different domains, it becomes more and more important to clearly understand their limitations. Specifically, understanding GANs' capability across the full spectrum of spatial frequencies, i.e. beyond the low-frequency dominant spectrum of natural images, is critical for assessing the reliability of GAN generated data in any detail-sensitive application (e.g. denoising, filling and super-resolution in medical and satellite images). In this paper, we show that the ability of convolutional GANs to learn a distribution is significantly affected by the spatial frequency of the underlying carrier signal, that is, GANs have a bias against learning high spatial frequencies. Crucially, we show that this bias is not merely a result of the scarcity of high frequencies in natural images, rather, it is a systemic bias hindering the learning of high frequencies regardless of their prominence in a dataset. Furthermore, we explain why large-scale GANs' ability to generate fine details on natural images does not exclude them from the adverse effects of this bias. Finally, we propose a method for manipulating this bias with minimal computational overhead. This method can be used to explicitly direct computational resources towards any specific spatial frequency of interest in a dataset, extending the flexibility of GANs.

Bingchen Liu, Kunpeng Song, Yizhe Zhu et al.

Focusing on text-to-image (T2I) generation, we propose Text and Image Mutual-Translation Adversarial Networks (TIME), a lightweight but effective model that jointly learns a T2I generator G and an image captioning discriminator D under the Generative Adversarial Network framework. While previous methods tackle the T2I problem as a uni-directional task and use pre-trained language models to enforce the image--text consistency, TIME requires neither extra modules nor pre-training. We show that the performance of G can be boosted substantially by training it jointly with D as a language model. Specifically, we adopt Transformers to model the cross-modal connections between the image features and word embeddings, and design an annealing conditional hinge loss that dynamically balances the adversarial learning. In our experiments, TIME achieves state-of-the-art (SOTA) performance on the CUB and MS-COCO dataset (Inception Score of 4.91 and Fréchet Inception Distance of 14.3 on CUB), and shows promising performance on MS-COCO on image captioning and downstream vision-language tasks.

Bingchen Liu, Kunpeng Song, Ahmed Elgammal

We propose a new approach for synthesizing fully detailed art-stylized images from sketches. Given a sketch, with no semantic tagging, and a reference image of a specific style, the model can synthesize meaningful details with colors and textures. The model consists of three modules designed explicitly for better artistic style capturing and generation. Based on a GAN framework, a dual-masked mechanism is introduced to enforce the content constraints (from the sketch), and a feature-map transformation technique is developed to strengthen the style consistency (to the reference image). Finally, an inverse procedure of instance-normalization is proposed to disentangle the style and content information, therefore yields better synthesis performance. Experiments demonstrate a significant qualitative and quantitative boost over baselines based on previous state-of-the-art techniques, adopted for the proposed process.

Shijie Geng, Ji Zhang, Hang Zhang et al.

We present a simple method that achieves unexpectedly superior performance for Complex Reasoning involved Visual Question Answering. Our solution collects statistical features from high-frequency words of all the questions asked about an image and use them as accurate knowledge for answering further questions of the same image. We are fully aware that this setting is not ubiquitously applicable, and in a more common setting one should assume the questions are asked separately and they cannot be gathered to obtain a knowledge base. Nonetheless, we use this method as an evidence to demonstrate our observation that the bottleneck effect is more severe on the feature extraction part than it is on the knowledge reasoning part. We show significant gaps when using the same reasoning model with 1) ground-truth features; 2) statistical features; 3) detected features from completely learned detectors, and analyze what these gaps mean to researches on visual reasoning topics. Our model with the statistical features achieves the 2nd place in the GQA Challenge 2019.

Bingchen Liu, Yizhe Zhu, Zuohui Fu et al.

Exploring the potential of GANs for unsupervised disentanglement learning, this paper proposes a novel GAN-based disentanglement framework with One-Hot Sampling and Orthogonal Regularization (OOGAN). While previous works mostly attempt to tackle disentanglement learning through VAE and seek to implicitly minimize the Total Correlation (TC) objective with various sorts of approximation methods, we show that GANs have a natural advantage in disentangling with an alternating latent variable (noise) sampling method that is straightforward and robust. Furthermore, we provide a brand-new perspective on designing the structure of the generator and discriminator, demonstrating that a minor structural change and an orthogonal regularization on model weights entails an improved disentanglement. Instead of experimenting on simple toy datasets, we conduct experiments on higher-resolution images and show that OOGAN greatly pushes the boundary of unsupervised disentanglement.

Yizhe Zhu, Jianwen Xie, Bingchen Liu et al.

We investigate learning feature-to-feature translator networks by alternating back-propagation as a general-purpose solution to zero-shot learning (ZSL) problems. It is a generative model-based ZSL framework. In contrast to models based on generative adversarial networks (GAN) or variational autoencoders (VAE) that require auxiliary networks to assist the training, our model consists of a single conditional generator that maps class-level semantic features and Gaussian white noise vector accounting for instance-level latent factors to visual features, and is trained by maximum likelihood estimation. The training process is a simple yet effective alternating back-propagation process that iterates the following two steps: (i) the inferential back-propagation to infer the latent factors of each observed example, and (ii) the learning back-propagation to update the model parameters. We show that, with slight modifications, our model is capable of learning from incomplete visual features for ZSL. We conduct extensive comparisons with existing generative ZSL methods on five benchmarks, demonstrating the superiority of our method in not only ZSL performance but also convergence speed and computational cost. Specifically, our model outperforms the existing state-of-the-art methods by a remarkable margin up to 3.1% and 4.0% in ZSL and generalized ZSL settings, respectively.

Ji Zhang, Kevin J. Shih, Ahmed Elgammal et al.

Most scene graph parsers use a two-stage pipeline to detect visual relationships: the first stage detects entities, and the second predicts the predicate for each entity pair using a softmax distribution. We find that such pipelines, trained with only a cross entropy loss over predicate classes, suffer from two common errors. The first, Entity Instance Confusion, occurs when the model confuses multiple instances of the same type of entity (e.g. multiple cups). The second, Proximal Relationship Ambiguity, arises when multiple subject-predicate-object triplets appear in close proximity with the same predicate, and the model struggles to infer the correct subject-object pairings (e.g. mis-pairing musicians and their instruments). We propose a set of contrastive loss formulations that specifically target these types of errors within the scene graph parsing problem, collectively termed the Graphical Contrastive Losses. These losses explicitly force the model to disambiguate related and unrelated instances through margin constraints specific to each type of confusion. We further construct a relationship detector, called RelDN, using the aforementioned pipeline to demonstrate the efficacy of our proposed losses. Our model outperforms the winning method of the OpenImages Relationship Detection Challenge by 4.7\% (16.5\% relative) on the test set. We also show improved results over the best previous methods on the Visual Genome and Visual Relationship Detection datasets.

Yizhe Zhu, Jianwen Xie, Zhiqiang Tang et al.

Zero-shot learning extends the conventional object classification to the unseen class recognition by introducing semantic representations of classes. Existing approaches predominantly focus on learning the proper mapping function for visual-semantic embedding, while neglecting the effect of learning discriminative visual features. In this paper, we study the significance of the discriminative region localization. We propose a semantic-guided multi-attention localization model, which automatically discovers the most discriminative parts of objects for zero-shot learning without any human annotations. Our model jointly learns cooperative global and local features from the whole object as well as the detected parts to categorize objects based on semantic descriptions. Moreover, with the joint supervision of embedding softmax loss and class-center triplet loss, the model is encouraged to learn features with high inter-class dispersion and intra-class compactness. Through comprehensive experiments on three widely used zero-shot learning benchmarks, we show the efficacy of the multi-attention localization and our proposed approach improves the state-of-the-art results by a considerable margin.

Ji Zhang, Kevin Shih, Andrew Tao et al.

We propose an efficient and interpretable scene graph generator. We consider three types of features: visual, spatial and semantic, and we use a late fusion strategy such that each feature's contribution can be explicitly investigated. We study the key factors about these features that have the most impact on the performance, and also visualize the learned visual features for relationships and investigate the efficacy of our model. We won the champion of the OpenImages Visual Relationship Detection Challenge on Kaggle, where we outperform the 2nd place by 5\% (20\% relatively). We believe an accurate scene graph generator is a fundamental stepping stone for higher-level vision-language tasks such as image captioning and visual QA, since it provides a semantic, structured comprehension of an image that is beyond pixels and objects.

Ji Zhang, Kevin Shih, Andrew Tao et al.

This article describes the model we built that achieved 1st place in the OpenImage Visual Relationship Detection Challenge on Kaggle. Three key factors contribute the most to our success: 1) language bias is a powerful baseline for this task. We build the empirical distribution $P(predicate|subject,object)$ in the training set and directly use that in testing. This baseline achieved the 2nd place when submitted; 2) spatial features are as important as visual features, especially for spatial relationships such as "under" and "inside of"; 3) It is a very effective way to fuse different features by first building separate modules for each of them, then adding their output logits before the final softmax layer. We show in ablation study that each factor can improve the performance to a non-trivial extent, and the model reaches optimal when all of them are combined.

Mahyar Khayatkhoei, Ahmed Elgammal, Maneesh Singh

Natural images may lie on a union of disjoint manifolds rather than one globally connected manifold, and this can cause several difficulties for the training of common Generative Adversarial Networks (GANs). In this work, we first show that single generator GANs are unable to correctly model a distribution supported on a disconnected manifold, and investigate how sample quality, mode dropping and local convergence are affected by this. Next, we show how using a collection of generators can address this problem, providing new insights into the success of such multi-generator GANs. Finally, we explain the serious issues caused by considering a fixed prior over the collection of generators and propose a novel approach for learning the prior and inferring the necessary number of generators without any supervision. Our proposed modifications can be applied on top of any other GAN model to enable learning of distributions supported on disconnected manifolds. We conduct several experiments to illustrate the aforementioned shortcoming of GANs, its consequences in practice, and the effectiveness of our proposed modifications in alleviating these issues.

Ji Zhang, Yannis Kalantidis, Marcus Rohrbach et al.

Large scale visual understanding is challenging, as it requires a model to handle the widely-spread and imbalanced distribution of <subject, relation, object> triples. In real-world scenarios with large numbers of objects and relations, some are seen very commonly while others are barely seen. We develop a new relationship detection model that embeds objects and relations into two vector spaces where both discriminative capability and semantic affinity are preserved. We learn both a visual and a semantic module that map features from the two modalities into a shared space, where matched pairs of features have to discriminate against those unmatched, but also maintain close distances to semantically similar ones. Benefiting from that, our model can achieve superior performance even when the visual entity categories scale up to more than 80,000, with extremely skewed class distribution. We demonstrate the efficacy of our model on a large and imbalanced benchmark based of Visual Genome that comprises 53,000+ objects and 29,000+ relations, a scale at which no previous work has ever been evaluated at. We show superiority of our model over carefully designed baselines on the original Visual Genome dataset with 80,000+ categories. We also show state-of-the-art performance on the VRD dataset and the scene graph dataset which is a subset of Visual Genome with 200 categories.

Ahmed Elgammal, Marian Mazzone, Bingchen Liu et al.

How does the machine classify styles in art? And how does it relate to art historians' methods for analyzing style? Several studies have shown the ability of the machine to learn and predict style categories, such as Renaissance, Baroque, Impressionism, etc., from images of paintings. This implies that the machine can learn an internal representation encoding discriminative features through its visual analysis. However, such a representation is not necessarily interpretable. We conducted a comprehensive study of several of the state-of-the-art convolutional neural networks applied to the task of style classification on 77K images of paintings, and analyzed the learned representation through correlation analysis with concepts derived from art history. Surprisingly, the networks could place the works of art in a smooth temporal arrangement mainly based on learning style labels, without any a priori knowledge of time of creation, the historical time and context of styles, or relations between styles. The learned representations showed that there are few underlying factors that explain the visual variations of style in art. Some of these factors were found to correlate with style patterns suggested by Heinrich Wölfflin (1846-1945). The learned representations also consistently highlighted certain artists as the extreme distinctive representative of their styles, which quantitatively confirms art historian observations.

Yizhe Zhu, Mohamed Elhoseiny, Bingchen Liu et al.

Most existing zero-shot learning methods consider the problem as a visual semantic embedding one. Given the demonstrated capability of Generative Adversarial Networks(GANs) to generate images, we instead leverage GANs to imagine unseen categories from text descriptions and hence recognize novel classes with no examples being seen. Specifically, we propose a simple yet effective generative model that takes as input noisy text descriptions about an unseen class (e.g.Wikipedia articles) and generates synthesized visual features for this class. With added pseudo data, zero-shot learning is naturally converted to a traditional classification problem. Additionally, to preserve the inter-class discrimination of the generated features, a visual pivot regularization is proposed as an explicit supervision. Unlike previous methods using complex engineered regularizers, our approach can suppress the noise well without additional regularization. Empirically, we show that our method consistently outperforms the state of the art on the largest available benchmarks on Text-based Zero-shot Learning.

Ahmed Elgammal, Yan Kang, Milko Den Leeuw

This paper proposes a computational approach for analysis of strokes in line drawings by artists. We aim at developing an AI methodology that facilitates attribution of drawings of unknown authors in a way that is not easy to be deceived by forged art. The methodology used is based on quantifying the characteristics of individual strokes in drawings. We propose a novel algorithm for segmenting individual strokes. We designed and compared different hand-crafted and learned features for the task of quantifying stroke characteristics. We also propose and compare different classification methods at the drawing level. We experimented with a dataset of 300 digitized drawings with over 80 thousands strokes. The collection mainly consisted of drawings of Pablo Picasso, Henry Matisse, and Egon Schiele, besides a small number of representative works of other artists. The experiments shows that the proposed methodology can classify individual strokes with accuracy 70%-90%, and aggregate over drawings with accuracy above 80%, while being robust to be deceived by fakes (with accuracy 100% for detecting fakes in most settings).

Mohamed Elhoseiny, Yizhe Zhu, Han Zhang et al.

In this paper, we study learning visual classifiers from unstructured text descriptions at part precision with no training images. We propose a learning framework that is able to connect text terms to its relevant parts and suppress connections to non-visual text terms without any part-text annotations. For instance, this learning process enables terms like "beak" to be sparsely linked to the visual representation of parts like head, while reduces the effect of non-visual terms like "migrate" on classifier prediction. Images are encoded by a part-based CNN that detect bird parts and learn part-specific representation. Part-based visual classifiers are predicted from text descriptions of unseen visual classifiers to facilitate classification without training images (also known as zero-shot recognition). We performed our experiments on CUBirds 2011 dataset and improves the state-of-the-art text-based zero-shot recognition results from 34.7\% to 43.6\%. We also created large scale benchmarks on North American Bird Images augmented with text descriptions, where we also show that our approach outperforms existing methods. Our code, data, and models are publically available.

Yizhe Zhu, Ahmed Elgammal

The exponentially increasing use of moving platforms for video capture introduces the urgent need to develop the general background subtraction algorithms with the capability to deal with the moving background. In this paper, we propose a multilayer-based framework for online background subtraction for videos captured by moving cameras. Unlike the previous treatments of the problem, the proposed method is not restricted to binary segmentation of background and foreground, but formulates it as a multi-label segmentation problem by modeling multiple foreground objects in different layers when they appear simultaneously in the scene. We assign an independent processing layer to each foreground object, as well as the background, where both motion and appearance models are estimated, and a probability map is inferred using a Bayesian filtering framework. Finally, Multi-label Graph-cut on Markov Random Field is employed to perform pixel-wise labeling. Extensive evaluation results show that the proposed method outperforms state-of-the-art methods on challenging video sequences.

Ahmed Elgammal, Bingchen Liu, Mohamed Elhoseiny et al.

We propose a new system for generating art. The system generates art by looking at art and learning about style; and becomes creative by increasing the arousal potential of the generated art by deviating from the learned styles. We build over Generative Adversarial Networks (GAN), which have shown the ability to learn to generate novel images simulating a given distribution. We argue that such networks are limited in their ability to generate creative products in their original design. We propose modifications to its objective to make it capable of generating creative art by maximizing deviation from established styles and minimizing deviation from art distribution. We conducted experiments to compare the response of human subjects to the generated art with their response to art created by artists. The results show that human subjects could not distinguish art generated by the proposed system from art generated by contemporary artists and shown in top art fairs. Human subjects even rated the generated images higher on various scales.

Mohamed Elhoseiny, Ahmed Elgammal

We present the Overlapping Domain Cover (ODC) notion for kernel machines, as a set of overlapping subsets of the data that covers the entire training set and optimized to be spatially cohesive as possible. We show how this notion benefit the speed of local kernel machines for regression in terms of both speed while achieving while minimizing the prediction error. We propose an efficient ODC framework, which is applicable to various regression models and in particular reduces the complexity of Twin Gaussian Processes (TGP) regression from cubic to quadratic. Our notion is also applicable to several kernel methods (e.g., Gaussian Process Regression(GPR) and IWTGP regression, as shown in our experiments). We also theoretically justified the idea behind our method to improve local prediction by the overlapping cover. We validated and analyzed our method on three benchmark human pose estimation datasets and interesting findings are discussed.

Gabriel Moyà-Alcover, Ahmed Elgammal, Antoni Jaume-i-Capó et al.

The problem of detecting changes in a scene and segmenting the foreground from background is still challenging, despite previous work. Moreover, new RGBD capturing devices include depth cues, which could be incorporated to improve foreground segmentation. In this work, we present a new nonparametric approach where a unified model mixes the device multiple information cues. In order to unify all the device channel cues, a new probabilistic depth data model is also proposed where we show how handle the inaccurate data to improve foreground segmentation. A new RGBD video dataset is presented in order to introduce a new standard for comparison purposes of this kind of algorithms. Results show that the proposed approach can handle several practical situations and obtain good results in all cases.

Mohamed Elhoseiny, Scott Cohen, Walter Chang et al.

Motivated by the application of fact-level image understanding, we present an automatic method for data collection of structured visual facts from images with captions. Example structured facts include attributed objects (e.g., <flower, red>), actions (e.g., <baby, smile>), interactions (e.g., <man, walking, dog>), and positional information (e.g., <vase, on, table>). The collected annotations are in the form of fact-image pairs (e.g.,<man, walking, dog> and an image region containing this fact). With a language approach, the proposed method is able to collect hundreds of thousands of visual fact annotations with accuracy of 83% according to human judgment. Our method automatically collected more than 380,000 visual fact annotations and more than 110,000 unique visual facts from images with captions and localized them in images in less than one day of processing time on standard CPU platforms.

Babak Saleh, Ahmed Elgammal, Jacob Feldman

Deep artificial neural networks have made remarkable progress in different tasks in the field of computer vision. However, the empirical analysis of these models and investigation of their failure cases has received attention recently. In this work, we show that deep learning models cannot generalize to atypical images that are substantially different from training images. This is in contrast to the superior generalization ability of the visual system in the human brain. We focus on Convolutional Neural Networks (CNN) as the state-of-the-art models in object recognition and classification; investigate this problem in more detail, and hypothesize that training CNN models suffer from unstructured loss minimization. We propose computational models to improve the generalization capacity of CNNs by considering how typical a training image looks like. By conducting an extensive set of experiments we show that involving a typicality measure can improve the classification results on a new set of images by a large margin. More importantly, this significant improvement is achieved without fine-tuning the CNN model on the target image set.

Amr Bakry, Ahmed Elgammal

Speech recognition is a challenging problem. Due to the acoustic limitations, using visual information is essential for improving the recognition accuracy in real-life unconstraint situations. One common approach is to model the visual recognition as nonlinear optimization problem. Measuring the distances between visual units is essential for solving this problem. Embedding the visual units on a manifold and using manifold kernels is one way to measure these distances. This work is intended to evaluate the performance of several manifold kernels for solving the problem of visual speech recognition. We show the theory behind each kernel. We apply manifold kernel partial least squares framework to OuluVs and AvLetters databases, and show empirical comparison between all kernels. This framework provides convenient way to explore different kernels.

Chetan Tonde, Ahmed Elgammal

Learning the kernel functions used in kernel methods has been a vastly explored area in machine learning. It is now widely accepted that to obtain 'good' performance, learning a kernel function is the key challenge. In this work we focus on learning kernel representations for structured regression. We propose use of polynomials expansion of kernels, referred to as Schoenberg transforms and Gegenbaur transforms, which arise from the seminal result of Schoenberg (1938). These kernels can be thought of as polynomial combination of input features in a high dimensional reproducing kernel Hilbert space (RKHS). We learn kernels over input and output for structured data, such that, dependency between kernel features is maximized. We use Hilbert-Schmidt Independence Criterion (HSIC) to measure this. We also give an efficient, matrix decomposition-based algorithm to learn these kernel transformations, and demonstrate state-of-the-art results on several real-world datasets.

Praneeth Vepakomma, Chetan Tonde, Ahmed Elgammal

In our work, we propose a novel formulation for supervised dimensionality reduction based on a nonlinear dependency criterion called Statistical Distance Correlation, Szekely et. al. (2007). We propose an objective which is free of distributional assumptions on regression variables and regression model assumptions. Our proposed formulation is based on learning a low-dimensional feature representation $\mathbf{z}$, which maximizes the squared sum of Distance Correlations between low dimensional features $\mathbf{z}$ and response $y$, and also between features $\mathbf{z}$ and covariates $\mathbf{x}$. We propose a novel algorithm to optimize our proposed objective using the Generalized Minimization Maximizaiton method of \Parizi et. al. (2015). We show superior empirical results on multiple datasets proving the effectiveness of our proposed approach over several relevant state-of-the-art supervised dimensionality reduction methods.

Mohamed Elhoseiny, Ahmed Elgammal, Babak Saleh

People typically learn through exposure to visual concepts associated with linguistic descriptions. For instance, teaching visual object categories to children is often accompanied by descriptions in text or speech. In a machine learning context, these observations motivates us to ask whether this learning process could be computationally modeled to learn visual classifiers. More specifically, the main question of this work is how to utilize purely textual description of visual classes with no training images, to learn explicit visual classifiers for them. We propose and investigate two baseline formulations, based on regression and domain transfer, that predict a linear classifier. Then, we propose a new constrained optimization formulation that combines a regression function and a knowledge transfer function with additional constraints to predict the parameters of a linear classifier. We also propose a generic kernelized models where a kernel classifier is predicted in the form defined by the representer theorem. The kernelized models allow defining and utilizing any two RKHS (Reproducing Kernel Hilbert Space) kernel functions in the visual space and text space, respectively. We finally propose a kernel function between unstructured text descriptions that builds on distributional semantics, which shows an advantage in our setting and could be useful for other applications. We applied all the studied models to predict visual classifiers on two fine-grained and challenging categorization datasets (CU Birds and Flower Datasets), and the results indicate successful predictions of our final model over several baselines that we designed.

Babak Saleh, Ahmed Elgammal, Jacob Feldman et al.

The human visual system can spot an abnormal image, and reason about what makes it strange. This task has not received enough attention in computer vision. In this paper we study various types of atypicalities in images in a more comprehensive way than has been done before. We propose a new dataset of abnormal images showing a wide range of atypicalities. We design human subject experiments to discover a coarse taxonomy of the reasons for abnormality. Our experiments reveal three major categories of abnormality: object-centric, scene-centric, and contextual. Based on this taxonomy, we propose a comprehensive computational model that can predict all different types of abnormality in images and outperform prior arts in abnormality recognition.

Mohamed Elhoseiny, Jingen Liu, Hui Cheng et al.

We propose a new zero-shot Event Detection method by Multi-modal Distributional Semantic embedding of videos. Our model embeds object and action concepts as well as other available modalities from videos into a distributional semantic space. To our knowledge, this is the first Zero-Shot event detection model that is built on top of distributional semantics and extends it in the following directions: (a) semantic embedding of multimodal information in videos (with focus on the visual modalities), (b) automatically determining relevance of concepts/attributes to a free text query, which could be useful for other applications, and (c) retrieving videos by free text event query (e.g., "changing a vehicle tire") based on their content. We embed videos into a distributional semantic space and then measure the similarity between videos and the event query in a free text form. We validated our method on the large TRECVID MED (Multimedia Event Detection) challenge. Using only the event title as a query, our method outperformed the state-of-the-art that uses big descriptions from 12.6% to 13.5% with MAP metric and 0.73 to 0.83 with ROC-AUC metric. It is also an order of magnitude faster.

Mohamed Elhoseiny, Tarek El-Gaaly, Amr Bakry et al.

In the task of Object Recognition, there exists a dichotomy between the categorization of objects and estimating object pose, where the former necessitates a view-invariant representation, while the latter requires a representation capable of capturing pose information over different categories of objects. With the rise of deep architectures, the prime focus has been on object category recognition. Deep learning methods have achieved wide success in this task. In contrast, object pose regression using these approaches has received relatively much less attention. In this paper we show how deep architectures, specifically Convolutional Neural Networks (CNN), can be adapted to the task of simultaneous categorization and pose estimation of objects. We investigate and analyze the layers of various CNN models and extensively compare between them with the goal of discovering how the layers of distributed representations of CNNs represent object pose information and how this contradicts with object category representations. We extensively experiment on two recent large and challenging multi-view datasets. Our models achieve better than state-of-the-art performance on both datasets.

Mohamed Elhoseiny, Scott Cohen, Walter Chang et al.

We study scalable and uniform understanding of facts in images. Existing visual recognition systems are typically modeled differently for each fact type such as objects, actions, and interactions. We propose a setting where all these facts can be modeled simultaneously with a capacity to understand unbounded number of facts in a structured way. The training data comes as structured facts in images, including (1) objects (e.g., $<$boy$>$), (2) attributes (e.g., $<$boy, tall$>$), (3) actions (e.g., $<$boy, playing$>$), and (4) interactions (e.g., $<$boy, riding, a horse $>$). Each fact has a semantic language view (e.g., $<$ boy, playing$>$) and a visual view (an image with this fact). We show that learning visual facts in a structured way enables not only a uniform but also generalizable visual understanding. We propose and investigate recent and strong approaches from the multiview learning literature and also introduce two learning representation models as potential baselines. We applied the investigated methods on several datasets that we augmented with structured facts and a large scale dataset of more than 202,000 facts and 814,000 images. Our experiments show the advantage of relating facts by the structure by the proposed models compared to the designed baselines on bidirectional fact retrieval.

Amr Bakry, Mohamed Elhoseiny, Tarek El-Gaaly et al.

This paper is focused on studying the view-manifold structure in the feature spaces implied by the different layers of Convolutional Neural Networks (CNN). There are several questions that this paper aims to answer: Does the learned CNN representation achieve viewpoint invariance? How does it achieve viewpoint invariance? Is it achieved by collapsing the view manifolds, or separating them while preserving them? At which layer is view invariance achieved? How can the structure of the view manifold at each layer of a deep convolutional neural network be quantified experimentally? How does fine-tuning of a pre-trained CNN on a multi-view dataset affect the representation at each layer of the network? In order to answer these questions we propose a methodology to quantify the deformation and degeneracy of view manifolds in CNN layers. We apply this methodology and report interesting results in this paper that answer the aforementioned questions.

Mohamed Elhoseiny, Ahmed Elgammal, Babak Saleh

In this paper we propose a framework for predicting kernelized classifiers in the visual domain for categories with no training images where the knowledge comes from textual description about these categories. Through our optimization framework, the proposed approach is capable of embedding the class-level knowledge from the text domain as kernel classifiers in the visual domain. We also proposed a distributional semantic kernel between text descriptions which is shown to be effective in our setting. The proposed framework is not restricted to textual descriptions, and can also be applied to other forms knowledge representations. Our approach was applied for the challenging task of zero-shot learning of fine-grained categories from text descriptions of these categories.

Ahmed Elgammal, Babak Saleh

Can we develop a computer algorithm that assesses the creativity of a painting given its context within art history? This paper proposes a novel computational framework for assessing the creativity of creative products, such as paintings, sculptures, poetry, etc. We use the most common definition of creativity, which emphasizes the originality of the product and its influential value. The proposed computational framework is based on constructing a network between creative products and using this network to infer about the originality and influence of its nodes. Through a series of transformations, we construct a Creativity Implication Network. We show that inference about creativity in this network reduces to a variant of network centrality problems which can be solved efficiently. We apply the proposed framework to the task of quantifying creativity of paintings (and sculptures). We experimented on two datasets with over 62K paintings to illustrate the behavior of the proposed framework. We also propose a methodology for quantitatively validating the results of the proposed algorithm, which we call the "time machine experiment".

Babak Saleh, Ahmed Elgammal

In the past few years, the number of fine-art collections that are digitized and publicly available has been growing rapidly. With the availability of such large collections of digitized artworks comes the need to develop multimedia systems to archive and retrieve this pool of data. Measuring the visual similarity between artistic items is an essential step for such multimedia systems, which can benefit more high-level multimedia tasks. In order to model this similarity between paintings, we should extract the appropriate visual features for paintings and find out the best approach to learn the similarity metric based on these features. We investigate a comprehensive list of visual features and metric learning approaches to learn an optimized similarity measure between paintings. We develop a machine that is able to make aesthetic-related semantic-level judgments, such as predicting a painting's style, genre, and artist, as well as providing similarity measures optimized based on the knowledge available in the domain of art historical interpretation. Our experiments show the value of using this similarity measure for the aforementioned prediction tasks.

Haopeng Zhang, Tarek El-Gaaly, Ahmed Elgammal et al.

Due to large variations in shape, appearance, and viewing conditions, object recognition is a key precursory challenge in the fields of object manipulation and robotic/AI visual reasoning in general. Recognizing object categories, particular instances of objects and viewpoints/poses of objects are three critical subproblems robots must solve in order to accurately grasp/manipulate objects and reason about their environments. Multi-view images of the same object lie on intrinsic low-dimensional manifolds in descriptor spaces (e.g. visual/depth descriptor spaces). These object manifolds share the same topology despite being geometrically different. Each object manifold can be represented as a deformed version of a unified manifold. The object manifolds can thus be parameterized by its homeomorphic mapping/reconstruction from the unified manifold. In this work, we develop a novel framework to jointly solve the three challenging recognition sub-problems, by explicitly modeling the deformations of object manifolds and factorizing it in a view-invariant space for recognition. We perform extensive experiments on several challenging datasets and achieve state-of-the-art results.

Sheng Huang, Mohamed Elhoseiny, Ahmed Elgammal et al.

We present a novel attribute learning framework named Hypergraph-based Attribute Predictor (HAP). In HAP, a hypergraph is leveraged to depict the attribute relations in the data. Then the attribute prediction problem is casted as a regularized hypergraph cut problem in which HAP jointly learns a collection of attribute projections from the feature space to a hypergraph embedding space aligned with the attribute space. The learned projections directly act as attribute classifiers (linear and kernelized). This formulation leads to a very efficient approach. By considering our model as a multi-graph cut task, our framework can flexibly incorporate other available information, in particular class label. We apply our approach to attribute prediction, Zero-shot and $N$-shot learning tasks. The results on AWA, USAA and CUB databases demonstrate the value of our methods in comparison with the state-of-the-art approaches.

Babak Saleh, Ali Farhadi, Ahmed Elgammal

When describing images, humans tend not to talk about the obvious, but rather mention what they find interesting. We argue that abnormalities and deviations from typicalities are among the most important components that form what is worth mentioning. In this paper we introduce the abnormality detection as a recognition problem and show how to model typicalities and, consequently, meaningful deviations from prototypical properties of categories. Our model can recognize abnormalities and report the main reasons of any recognized abnormality. We introduce the abnormality detection dataset and show interesting results on how to reason about abnormalities.

Sheng Huang, Ahmed Elgammal, Dan Yang

Hypergraph is a powerful representation in several computer vision, machine learning and pattern recognition problems. In the last decade, many researchers have been keen to develop different hypergraph models. In contrast, no much attention has been paid to the design of hyperedge weights. However, many studies on pairwise graphs show that the choice of edge weight can significantly influence the performances of such graph algorithms. We argue that this also applies to hypegraphs. In this paper, we empirically discuss the influence of hyperedge weight on hypegraph learning via proposing three novel hyperedge weights from the perspectives of geometry, multivariate statistical analysis and linear regression. Extensive experiments on ORL, COIL20, JAFFE, Sheffield, Scene15 and Caltech256 databases verify our hypothesis. Similar to graph learning, several representative hyperedge weighting schemes can be concluded by our experimental studies. Moreover, the experiments also demonstrate that the combinations of such weighting schemes and conventional hypergraph models can get very promising classification and clustering performances in comparison with some recent state-of-the-art algorithms.

Emily L. Spratt, Ahmed Elgammal

In part one of the Critique of Judgment, Immanuel Kant wrote that "the judgment of taste...is not a cognitive judgment, and so not logical, but is aesthetic."\cite{Kant} While the condition of aesthetic discernment has long been the subject of philosophical discourse, the role of the arbiters of that judgment has more often been assumed than questioned. The art historian, critic, connoisseur, and curator have long held the esteemed position of the aesthetic judge, their training, instinct, and eye part of the inimitable subjective processes that Kant described as occurring upon artistic evaluation. Although the concept of intangible knowledge in regard to aesthetic theory has been much explored, little discussion has arisen in response to the development of new types of artificial intelligence as a challenge to the seemingly ineffable abilities of the human observer. This paper examines the developments in the field of computer vision analysis of paintings from canonical movements with the history of Western art and the reaction of art historians to the application of this technology in the field. Through an investigation of the ethical consequences of this innovative technology, the unquestioned authority of the art expert is challenged and the subjective nature of aesthetic judgment is brought to philosophical scrutiny once again.

Mohamed Elhoseiny, Ahmed Elgammal

There has been a growing interest in mutual information measures due to their wide range of applications in Machine Learning and Computer Vision. In this paper, we present a generalized structured regression framework based on Shama-Mittal divergence, a relative entropy measure, which is introduced to the Machine Learning community in this work. Sharma-Mittal (SM) divergence is a generalized mutual information measure for the widely used Rényi, Tsallis, Bhattacharyya, and Kullback-Leibler (KL) relative entropies. Specifically, we study Sharma-Mittal divergence as a cost function in the context of the Twin Gaussian Processes (TGP)~\citep{Bo:2010}, which generalizes over the KL-divergence without computational penalty. We show interesting properties of Sharma-Mittal TGP (SMTGP) through a theoretical analysis, which covers missing insights in the traditional TGP formulation. However, we generalize this theory based on SM-divergence instead of KL-divergence which is a special case. Experimentally, we evaluated the proposed SMTGP framework on several datasets. The results show that SMTGP reaches better predictions than KL-based TGP, since it offers a bigger class of models through its parameters that we learn from the data.

Babak Saleh, Kanako Abe, Ravneet Singh Arora et al.

Considering the huge amount of art pieces that exist, there is valuable information to be discovered. Examining a painting, an expert can determine its style, genre, and the time period that the painting belongs. One important task for art historians is to find influences and connections between artists. Is influence a task that a computer can measure? The contribution of this paper is in exploring the problem of computer-automated suggestion of influences between artists, a problem that was not addressed before in a general setting. We first present a comparative study of different classification methodologies for the task of fine-art style classification. A two-level comparative study is performed for this classification problem. The first level reviews the performance of discriminative vs. generative models, while the second level touches the features aspect of the paintings and compares semantic-level features vs. low-level and intermediate-level features present in the painting. Then, we investigate the question "Who influenced this artist?" by looking at his masterpieces and comparing them to others. We pose this interesting question as a knowledge discovery problem. For this purpose, we investigated several painting-similarity and artist-similarity measures. As a result, we provide a visualization of artists (Map of Artists) based on the similarity between their works