Canada

Face retouching is to beautify a face image, while preserving the image content as much as possible. It is a promising yet challenging task to remove face imperfections and fill with normal skin. Generic image enhancement methods are hampered by the lack of imperfection localization, which often results in incomplete removal of blemishes at large scales. To address this issue, we propose a transformer-based approach, RetouchFormer, which simultaneously identify imperfections and synthesize realistic content in the corresponding regions. Specifically, we learn a latent dictionary to capture the clean face priors, and predict the imperfection regions via a reconstruction-oriented localization module. Also based on this, we can realize face retouching by explicitly suppressing imperfections in our selective self-attention computation, such that local content will be synthesized from normal skin. On the other hand, multi-scale feature tokens lead to increased flexibility in dealing with the imperfections at various scales. The design elements bring greater effectiveness and efficiency. RetouchFormer outperforms the advanced face retouching methods and synthesizes clean face images with high fidelity in our list of extensive experiments performed.

AAAI 2024

RetouchFormer: Semi-supervised High-Quality Face Retouching Transformer with Prior-Based Selective Self-Attention | VIDEO

computer vison; face; image synthesis;

technical paper

We are pleased to announce the Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24), which will be held in Vancouver, British Columbia at the Vancouver Convention Centre – West Building from 20-27 February, 2024.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-24 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

We expect for AAAI-24 to be an in-person conference – one author of all accepted papers will be expected to present work in person unless there are exceptional circumstances that prevent this.<br><br><br><br>

In order to access the AAAI-24 event page you need to register [here](https://aaai.org/aaai-conference/registration/)

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. 

This paper delves into the realm of stochastic optimization for compositional minimax optimization—a pivotal challenge across various machine learning domains, including deep AUC and reinforcement learning policy evaluation. Despite its significance, the problem of compositional minimax optimization is still under-explored. Adding to the complexity, current methods of compositional minimax optimization are plagued by sub-optimal complexities or heavy reliance on sizable batch sizes. To respond to these constraints, this paper introduces a novel method, called Nested STOchastic Recursive Momentum (NSTORM), which can achieve the optimal sample complexity and obtain the nearly accuracy solution, matching the existing minimax methods. We also demonstrate that NSTORM can achieve the same sample complexity under the Polyak-Lojasiewicz (PL)-condition—an insightful extension of its capabilities. Yet, NSTORM encounters an issue with its requirement for low learning rates, potentially constraining its real-world applicability in machine learning. To overcome this hurdle, we present ADAptive NSTORM (ADA-NSTORM) with adaptive learning rates. We demonstrate that ADA-NSTORM can achieve the same sample complexity but the experimental results show its more effectiveness. All the proposed complexities indicate that our proposed methods can match lower bounds to existing minimax optimizations, without requiring a large batch size in each iteration. Extensive experiments support the efficiency of our proposed methods.

Faster Stochastic Variance Reduction Methods for Compositional MiniMax Optimization | VIDEO

Dance is generally considered to be complex for most people as it requires decent performances of numerous body motions and proper responses to the musical content and rhythm. Studies on automatic dance performance assessment could help people improve their sensorimotor skills and promote research in many fields, including human motion analysis and motion generation. Recent works on dance performance assessment usually evaluate simple dance motions with a single task - estimating final performance scores. In this paper, we propose ImperialDance: multi-task dance performance assessment via text prompting that solves three related tasks - (i) dance vocabulary recognition, (ii) dance performance scoring and (iii) dance rhythm evaluation. In the pre-training phase, we contrastively learn the primitive-based features of complex dance motion and music using the InfoNCE loss. For the downstream task, we propose a transformer-based text prompter to perform multi-task evaluations for the three proposed assessment tasks. Also, we built a multimodal dance-music dataset named ImperialDance. The novelty of our ImperialDance is that it contains dance motions for diverse expertise levels and a significant amount of repeating dance sequences for the same choreography to keep track of the dance performance progression. Qualitative results show that our pre-trained feature representation could cluster dance pieces for different dance genres, choreographies, expertise levels and primitives, which generalizes well on both ours and other dance-music datasets. The downstream experiments demonstrate the robustness and improvement of our method over several ablations and baselines across all three tasks, as well as users' dance level progression monitoring.

DanceMVP: Self-Supervised Learning for Multi-Task Primitive-Based Dance Performance Assessment via Transformer Text Prompting

In this paper, we study a fundamental problem in submodular optimization known as sequential submodular maximization. The primary objective of this problem is to select and rank a sequence of items to optimize a group of submodular functions.
The existing research on this problem has predominantly concentrated on the monotone setting, assuming that the submodular functions are non-decreasing. However, in various real-world scenarios, like diversity-aware recommendation systems, adding items to an existing set might negatively impact the overall utility. In response, we propose to study this problem with non-monotone submodular functions and develop approximation algorithms for both flexible and fixed length constraints, as well as a special case with identical utility functions. The empirical evaluations further validate the effectiveness of our proposed algorithms in the domain of video recommendations.

Non-monotone Sequential Submodular Maximization

Learning-based point cloud completion tasks have shown potential in various critical tasks, such as object detection, assignment, and registration. However, accurately and efficiently quantifying the shape error between the predicted point clouds generated by networks and the ground truth remains challenging. While EMD-based loss functions excel in shape detail and perceived density distribution, their approach can only yield results with significant discrepancies from the actual EMD within a tolerable training time.
To address these challenges, we first propose the initial price based on the auction algorithm, reducing the number of iterations required for the algorithm while ensuring the correctness of the assignment results. We then introduce an algorithm to compute the initial price through a successive shortest path and the Euclidean information between its nodes. Finally, we adopt a series of optimization strategies to speed up the algorithm and offer an EMD approximation scheme for point cloud problems that balances time loss and computational accuracy based on point cloud data characteristics.
Our experimental results confirm that our algorithm achieves the smallest gap with the real EMD within an acceptable time range and yields the best results in end-to-end training.

A Computation-Aware Shape Loss Function for Point Cloud Completion

Conversational Speech Synthesis (CSS) aims to accurately express an utterance with the appropriate prosody and emotional inflection within a conversational setting. While recognising the significance of CSS task, the prior studies have not thoroughly investigated the emotional expressiveness problems due to the scarcity of emotional conversational datasets and the difficulty of stateful emotion modeling. In this paper, we propose a novel emotional CSS model, termed ECSS, that includes two main components: 1) to enhance emotion understanding, we introduce a heterogeneous graph-based emotional context modeling mechanism, which takes the multi-source dialogue history as input to model the dialogue context and learn the emotion cues from the context; 2) to achieve emotion rendering, we employ a contrastive learning-based emotion renderer module to infer the accurate emotion style for the target utterance. To address the issue of data scarcity, we meticulously create emotional labels in terms of category and intensity, and annotate additional emotional information on the existing conversational dataset (DailyTalk). Both objective and subjective evaluations suggest that our model outperforms the baseline models in understanding and rendering emotions. These evaluations also underscore the importance of comprehensive emotional annotations.  Code and audio samples can be found at: https://github.com/walker-hyf/ECSS.

Emotion Rendering for Conversational Speech Synthesis with Heterogeneous Graph-Based Context Modeling | VIDEO

Deep neural networks have shown remarkable performance in image classification. However, their performance significantly deteriorates with corrupted input data. Domain generalization methods have been proposed to train robust models against out-of-distribution data. Data augmentation in the frequency domain is one of such approaches that enable a model to learn phase features to establish domain-invariant representations. This approach changes the amplitudes of the input data while preserving the phases. However, using fixed phases leads to susceptibility to phase fluctuations because amplitudes and phase fluctuations commonly occur in out-of-distribution. In this study, to address this problem, we introduce an approach using finite variation of the phases of input data rather than maintaining fixed phases. Based on the assumption that the degree of domain-invariant features varies for each phase, we propose a method to distinguish phases based on this degree. In addition, we propose a method called vital phase augmentation (VIPAug) that applies the variation to the phases differently according to the degree of domain-invariant features of given phases. The model depends more on the vital phases that contain more domain-invariant features for attaining robustness to amplitude and phase fluctuations. We present experimental evaluations of our proposed approach, which exhibited improved performance for both clean and corrupted data. VIPAug achieved SOTA performance on the benchmark CIFAR-10 and CIFAR-100 datasets, as well as near-SOTA performance on the ImageNet-100 and ImageNet datasets.

Domain Generalization with Vital Phase Augmentation

In this paper, we study the two-level ski-rental problem, where a user needs to fulfill a sequence of demands for multiple items by choosing one of the three payment options: paying for the on-demand usage (i.e., rent), buying individual items (i.e., single purchase), and buying all the items (i.e., combo purchase). Without knowing future demands, the user aims to minimize the total cost (i.e., the sum of the rental, single purchase, and combo purchase costs) by balancing the trade-off between the expensive upfront costs (for purchase) and the potential future expenses (for rent). We first design a robust online algorithm (RDTSR) that offers a worst-case performance guarantee. While online algorithms are robust against the worst-case scenarios, they are often overly cautious and thus suffer a poor average performance in typical scenarios. On the other hand, Machine Learning (ML) algorithms typically show promising average performance in various applications but lack worst-case performance guarantees. To harness the benefits of both methods, we develop a learning-augmented algorithm (LADTSR) by integrating ML predictions into the robust online algorithm, which outperforms the robust online algorithm under accurate predictions while ensuring worst-case performance guarantees even when predictions are inaccurate. Finally, we conduct numerical experiments on both synthetic and real-world trace data to corroborate the effectiveness of our approach.

Learning-Augmented Online Algorithm for Two-Level Ski-Rental Problem

We develop a general algorithmic framework that allows us to obtain fixed-parameter tractability for computing smallest symbolic models that represent given data. Our framework applies to all ML model types that admit a certain extension property. By showing this extension property for decision trees, decision sets, decision lists, and binary decision diagrams, we obtain that minimizing these fundamental model types is fixed-parameter tractable. Our framework even applies to ensembles, which combine individual models by majority decision.

A General Theoretical Framework for Learning Smallest Interpretable Models

Visual Question Answering (VQA) is a fundamental task in computer vision and natural language process fields. Although the ``pre-training & finetuning'' learning paradigm significantly improves the VQA performance, the adversarial robustness of such a paradigm has not been explored. 
In this paper, we delve into a new problem: using a pre-trained multimodal source model to create adversarial image-text pairs and then transferring them to attack the target VQA models.
Correspondingly, we propose a novel VQAttack model, which can iteratively generate both image and text perturbations with the designed modules: the large language model (LLM)-enhanced image attack and the cross-modal joint attack module.
At each iteration, the LLM-enhanced image attack module first optimizes the latent representation-based loss to generate feature-level image perturbations. Then it incorporates an LLM to further enhance the image perturbations by optimizing the designed masked answer anti-recovery loss. 
The cross-modal joint attack module will be triggered at a specific iteration, which updates the image and text perturbations sequentially. Notably, the text perturbation updates are based on both the learned gradients in the word embedding space and word synonym-based substitution.
Experimental results on two VQA datasets with five validated models demonstrate the effectiveness of the proposed VOAttack in the transferable attack setting, compared with state-of-the-art baselines. This work also reveals a significant blind spot in the ``pre-training \& fine-tuning'' paradigm on VQA tasks.

VQAttack: Transferable Adversarial Attacks on Visual Question Answering via Pre-trained Models

The deployment of multi-stream fusion strategy on behavioral recognition from skeletal data can extract complementary features from different information streams and improve the recognition accuracy, but suffers from high model complexity and a large number of parameters. Besides, existing multi-stream methods using a fixed adjacency matrix homogenizes the model’s discrimination process across diverse actions, causing reduction of the actual lift for the multi-stream model. Finally, attention mechanisms are commonly applied to the multi-dimensional features, including spatial, temporal and channel dimensions. But their attention scores are typically fused in a concatenated manner, leading to the ignorance of the interrelation between joints in complex actions. To alleviate these issues, the Front-Rear dual Fusion Graph Convolutional Network (FRF-GCN) is proposed to provide a lightweight model based on skeletal data. Targeted adjacency matrices are also designed for different front fusion streams, allowing the model to focus on actions of varying magnitudes. Simultaneously, the mechanism of Spatial-Temporal-Channel Parallel Attention (STC-P), which processes attention in parallel and places greater emphasis on useful information, is proposed to further improve model’s performance. FRF-GCN demonstrates significant competitiveness compared to the current state-of-the-art methods on the NTU RGB+D, NTU RGB+D 120 and Kinetics-Skeleton 400 datasets. Our code is available at: https://github.com/sunbeam-kkt/FRF-GCN-master.

RetouchFormer: Semi-supervised High-Quality Face Retouching Transformer with Prior-Based Selective Self-Attention | VIDEO

Downloads

Next from AAAI 2024

Faster Stochastic Variance Reduction Methods for Compositional MiniMax Optimization | VIDEO

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES