AAAI 2024 论文列表

Speech emotions are crucial in human communication and are extensively used in fields like speech synthesis and natural language understanding. Most prior studies, such as speech emotion recognition, have categorized speech emotions into a fixed set of classes. Yet, emotions expressed in human speech are often complex, and categorizing them into predefined groups can be insufficient to adequately represent speech emotions. On the contrary, describing speech emotions directly by means of natural language may be a more effective approach. Regrettably, there are not many studies available that have focused on this direction. Therefore, this paper proposes a speech emotion captioning framework named SECap, aiming at effectively describing speech emotions using natural language. Owing to the impressive capabilities of large language models in language comprehension and text generation, SECap employs LLaMA as the text decoder to allow the production of coherent speech emotion captions. In addition, SECap leverages HuBERT as the audio encoder to extract general speech features and Q-Former as the Bridge-Net to provide LLaMA with emotion-related speech features. To accomplish this, Q-Former utilizes mutual information learning to disentangle emotion-related speech features and speech contents, while implementing contrastive learning to extract more emotion-related speech features. The results of objective and subjective evaluations demonstrate that: 1) the SECap framework outperforms the HTSAT-BART baseline in all objective evaluations; 2) SECap can generate high-quality speech emotion captions that attain performance on par with human annotators in subjective mean opinion score tests.

As one of the most popular machine learning tools in the field of unsupervised learning, clustering has been widely used in various practical applications. While numerous methods have been proposed for clustering, a commonly encountered issue is that the existing clustering methods rely heavily on local neighborhood information during the optimization process, which leads to suboptimal performance on real-world datasets. Besides, most existing clustering methods use Euclidean distances or densities to measure the similarity between data points. This could constrain the effectiveness of the algorithms for handling datasets with irregular patterns. Thus, a key challenge is how to effectively capture the global structural information in clustering instances to improve the clustering quality. In this paper, we propose a new clustering algorithm, called SEC. This algorithm uses the global structural information extracted from an encoding tree to guide the clustering optimization process. Based on the relation between data points in the instance, a sparse graph of the clustering instance can be constructed. By leveraging the sparse graph constructed, we propose an iterative encoding tree method, where hierarchical abstractions of the encoding tree are iteratively extracted as new clustering features to obtain better clustering results. To avoid the influence of easily misclustered data points located on the boundaries of the clustering partitions, which we call "fringe points", we propose an iterative pre-deletion and reassignment technique such that the algorithm can delete and reassign the "fringe points" to obtain more resilient and precise clustering results. Empirical experiments on both synthetic and real-world datasets demonstrate that our proposed algorithm outperforms state-of-the-art clustering methods and achieves better clustering performances. On average, the clustering accuracy (ACC) is increased by 1.7% and the normalized mutual information (NMI) by 7.9% compared with the current state-of-the-art (SOTA) algorithm on synthetic datasets. On real-world datasets, our method outperforms other clustering methods with an average increase of 12.3% in ACC and 5.2% in NMI, respectively.

The utilization of wavelet-based techniques in graph neural networks (GNNs) has gained considerable attention, particularly in the context of node classification. Although existing wavelet-based approaches have shown promise, they are constrained by their reliance on pre-defined wavelet filters, rendering them incapable of effectively adapting to signals that reside on graphs based on tasks at hand. Recent research endeavors address this issue through the introduction of a wavelet lifting transform. However, this technique necessitates the use of bipartite graphs, causing a transformation of the original graph structure into a bipartite configuration. This alteration of graph topology results in the generation of undesirable wavelet filters, thereby undermining the effectiveness of the method. In response to these challenges, we propose a novel simple and effective adaptive graph wavelet neural network (SEA-GWNN) class that employs the lifting scheme on arbitrary graph structures while upholding the original graph topology by leveraging multi-hop computation trees. A noteworthy aspect of the approach is the focus on local substructures represented as acyclic trees, wherein the lifting strategy is applied in a localized manner. This locally defined lifting scheme effectively combines high-pass and low-pass frequency information to enhance node representations. Furthermore, to reduce computing costs, we propose to decouple the higher- order lifting operators and induce them from the lower-order structures. Finally, we benchmark our model on several real- world datasets spanning four distinct categories, including citation networks, webpages, the film industry, and large-scale graphs and the experimental results showcase the efficacy of the proposed SEA-GWNN.

Rescaling the backpropagated gradient of contrastive loss has made significant progress in descriptor learning. However, current gradient modulation strategies have no regard for the varying distribution of global gradients, so they would suffer from changes in training phases or datasets. In this paper, we propose a dynamic gradient modulation, named SDGMNet, for contrastive local descriptor learning. The core of our method is formulating modulation functions with dynamically estimated statistical characteristics. Firstly, we introduce angle for distance measure after deep analysis on backpropagation of pair-wise loss. On this basis, auto-focus modulation is employed to moderate the impact of statistically uncommon individual pairs in stochastic gradient descent optimization; probabilistic margin cuts off the gradients of proportional triplets that have achieved enough optimization; power adjustment balances the total weights of negative pairs and positive pairs. Extensive experiments demonstrate that our novel descriptor surpasses previous state-of-the-art methods in several tasks including patch verification, retrieval, pose estimation, and 3D reconstruction.

Facial attribute editing has garnered significant attention, yet prevailing methods struggle with achieving precise attribute manipulation while preserving irrelevant details and controlling attribute styles. This challenge primarily arises from the strong correlations between different attributes and the interplay between attributes and identity. In this paper, we propose Semantic Disentangled GAN (SDGAN), a novel method addressing this challenge. SDGAN introduces two key concepts: a semantic disentanglement generator that assigns facial representations to distinct attribute-specific editing modules, enabling the decoupling of the facial attribute editing process, and a semantic mask alignment strategy that confines attribute editing to appropriate regions, thereby avoiding undesired modifications. Leveraging these concepts, SDGAN demonstrates accurate attribute editing and achieves high-quality attribute style manipulation through both latent-guided and reference-guided manners. We extensively evaluate our method on the CelebA-HQ database, providing both qualitative and quantitative analyses. Our results establish that SDGAN significantly outperforms state-of-the-art techniques, showcasing the effectiveness of our approach. To foster reproducibility and further research, we will provide the code for our method.

Nowadays, closed-set perception methods for autonomous driving perform well on datasets containing normal scenes. However, they still struggle to handle anomalies in the real world, such as unknown objects that have never been seen while training. The lack of public datasets to evaluate the model performance on anomaly and corner cases has hindered the development of reliable autonomous driving systems. Therefore, we propose a multimodal Synthetic Dataset for Anomaly and Corner case detection, called SDAC, which encompasses anomalies captured from multi-view cameras and the LiDAR sensor, providing a rich set of annotations for multiple mainstream perception tasks. SDAC is the first public dataset for autonomous driving that categorizes anomalies into object, scene, and scenario levels, allowing the evaluation under different anomalous conditions. Experiments show that closed-set models suffer significant performance drops on anomaly subsets in SDAC. Existing anomaly detection methods fail to achieve satisfactory performance, suggesting that anomaly detection remains a challenging problem. We anticipate that our SDAC dataset could foster the development of safe and reliable systems for autonomous driving.

In this paper, we introduce Segmentation-Driven Deformation Multi-View Stereo (SD-MVS), a method that can effectively tackle challenges in 3D reconstruction of textureless areas. We are the first to adopt the Segment Anything Model (SAM) to distinguish semantic instances in scenes and further leverage these constraints for pixelwise patch deformation on both matching cost and propagation. Concurrently, we propose a unique refinement strategy that combines spherical coordinates and gradient descent on normals and pixelwise search interval on depths, significantly improving the completeness of reconstructed 3D model. Furthermore, we adopt the Expectation-Maximization (EM) algorithm to alternately optimize the aggregate matching cost and hyperparameters, effectively mitigating the problem of parameters being excessively dependent on empirical tuning. Evaluations on the ETH3D high-resolution multi-view stereo benchmark and the Tanks and Temples dataset demonstrate that our method can achieve state-of-the-art results with less time consumption.

Recent real-time semantic segmentation methods usually adopt an additional semantic branch to pursue rich long-range context. However, the additional branch incurs undesirable computational overhead and slows inference speed. To eliminate this dilemma, we propose SCTNet, a single branch CNN with transformer semantic information for real-time segmentation. SCTNet enjoys the rich semantic representations of an inference-free semantic branch while retaining the high efficiency of lightweight single branch CNN. SCTNet utilizes a transformer as the training-only semantic branch considering its superb ability to extract long-range context. With the help of the proposed transformer-like CNN block CFBlock and the semantic information alignment module, SCTNet could capture the rich semantic information from the transformer branch in training. During the inference, only the single branch CNN needs to be deployed. We conduct extensive experiments on Cityscapes, ADE20K, and COCO-Stuff-10K, and the results show that our method achieves the new state-of-the-art performance. The code and model is available at https://github.com/xzz777/SCTNet.

In recent years, the task of learned point cloud compression has gained prominence. An important type of point cloud, LiDAR point cloud, is generated by spinning LiDAR on vehicles. This process results in numerous circular shapes and azimuthal angle invariance features within the point clouds. However, these two features have been largely overlooked by previous methodologies. In this paper, we introduce a model-agnostic method called Spherical-Coordinate-based learned Point cloud compression (SCP), designed to fully leverage the features of circular shapes and azimuthal angle invariance. Additionally, we propose a multi-level Octree for SCP to mitigate the reconstruction error for distant areas within the Spherical-coordinate-based Octree. SCP exhibits excellent universality, making it applicable to various learned point cloud compression techniques. Experimental results demonstrate that SCP surpasses previous state-of-the-art methods by up to 29.14% in point-to-point PSNR BD-Rate.

Contrastive learning has achieved great success in skeleton-based action recognition. However, most existing approaches encode the skeleton sequences as entangled spatiotemporal representations and confine the contrasts to the same level of representation. Instead, this paper introduces a novel contrastive learning framework, namely Spatiotemporal Clues Disentanglement Network (SCD-Net). Specifically, we integrate the decoupling module with a feature extractor to derive explicit clues from spatial and temporal domains respectively. As for the training of SCD-Net, with a constructed global anchor, we encourage the interaction between the anchor and extracted clues. Further, we propose a new masking strategy with structural constraints to strengthen the contextual associations, leveraging the latest development from masked image modelling into the proposed SCD-Net. We conduct extensive evaluations on the NTU-RGB+D (60&120) and PKU-MMD (I&II) datasets, covering various downstream tasks such as action recognition, action retrieval, transfer learning, and semi-supervised learning. The experimental results demonstrate the effectiveness of our method, which outperforms the existing state-of-the-art (SOTA) approaches significantly. Our code and supplementary material can be found at https://github.com/cong-wu/SCD-Net.

The recent neural surface reconstruction approaches using volume rendering have made much progress by achieving impressive surface reconstruction quality, but are still limited to dense and highly accurate posed views. To overcome such drawbacks, this paper pays special attention on the consistent surface reconstruction from sparse views with noisy camera poses. Unlike previous approaches, the key difference of this paper is to exploit the multi-view constraints directly from the explicit geometry of the neural surface, which can be used as effective regularization to jointly learn the neural surface and refine the camera poses. To build effective multi-view constraints, we introduce a fast differentiable on-surface intersection to generate on-surface points, and propose view-consistent losses on such differentiable points to regularize the neural surface learning. Based on this point, we propose a joint learning strategy, named SC-NeuS, to perform geometry-consistent surface reconstruction in an end-to-end manner. With extensive evaluation on public datasets, our SC-NeuS can achieve consistently better surface reconstruction results with fine-grained details than previous approaches, especially from sparse and noisy camera views. The source code is available at https://github.com/zouzx/sc-neus.git.

As a prominent parameter-efficient fine-tuning technique in NLP, prompt tuning is being explored its potential in computer vision. Typical methods for visual prompt tuning follow the sequential modeling paradigm stemming from NLP, which represents an input image as a flattened sequence of token embeddings and then learns a set of unordered parameterized tokens prefixed to the sequence representation as the visual prompts for task adaptation of large vision models. While such sequential modeling paradigm of visual prompt has shown great promise, there are two potential limitations. First, the learned visual prompts cannot model the underlying spatial relations in the input image, which is crucial for image encoding. Second, since all prompt tokens play the same role of prompting for all image tokens without distinction, it lacks the fine-grained prompting capability, i.e., individual prompting for different image tokens. In this work, we propose the Spatially Aligned-and-Adapted Visual Prompt model (SA^2VP), which learns a two-dimensional prompt token map with equal (or scaled) size to the image token map, thereby being able to spatially align with the image map. Each prompt token is designated to prompt knowledge only for the spatially corresponding image tokens. As a result, our model can conduct individual prompting for different image tokens in a fine-grained manner. Moreover, benefiting from the capability of preserving the spatial structure by the learned prompt token map, our SA^2VP is able to model the spatial relations in the input image, leading to more effective prompting. Extensive experiments on three challenging benchmarks for image classification demonstrate the superiority of our model over other state-of-the-art methods for visual prompt tuning. Code is available at https://github.com/tommy-xq/SA2VP.

Deep learning-based video super-resolution (VSR) networks have gained significant performance improvements in recent years. However, existing VSR networks can only support a fixed integer scale super-resolution task, and when we want to perform VSR at multiple scales, we need to train several models. This implementation certainly increases the consumption of computational and storage resources, which limits the application scenarios of VSR techniques. In this paper, we propose a novel Scale-adaptive Arbitrary-scale Video Super-Resolution network (SAVSR), which is the first work focusing on spatial VSR at arbitrary scales including both non-integer and asymmetric scales. We also present an omni-dimensional scale-attention convolution, which dynamically adapts according to the scale of the input to extract inter-frame features with stronger representational power. Moreover, the proposed spatio-temporal adaptive arbitrary-scale upsampling performs VSR tasks using both temporal features and scale information. And we design an iterative bi-directional architecture for implicit feature alignment. Experiments at various scales on the benchmark datasets show that the proposed SAVSR outperforms state-of-the-art (SOTA) methods at non-integer and asymmetric scales. The source code is available at https://github.com/Weepingchestnut/SAVSR.

Zero-shot object detection (ZSD) aims to localize and classify unseen objects without access to their training annotations. As a prevailing solution to ZSD, generation-based methods synthesize unseen visual features by taking seen features as reference and class semantic embeddings as guideline. Although previous works continuously improve the synthesis quality, they fail to consider the scale-varying nature of unseen objects. The generation process is preformed over a single scale of object features and thus lacks scale-diversity among synthesized features. In this paper, we reveal the scale-varying challenge in ZSD and propose a Scale-Aware Unseen Imagineer (SAUI) to lead the way of a novel scale-aware ZSD paradigm. To obtain multi-scale features of seen-class objects, we design a specialized coarse-to-fine extractor to capture features through multiple scale-views. To generate unseen features scale by scale, we innovate a Series-GAN synthesizer along with three scale-aware contrastive components to imagine separable, diverse and robust scale-wise unseen features. Extensive experiments on PASCAL VOC, COCO and DIOR datasets demonstrate SAUI's better performance in different scenarios, especially for scale-varying and small objects. Notably, SAUI achieves the new state-of-the art performance on COCO and DIOR.

Solvers for propositional satisfiability (SAT) effectively tackle hard optimization problems. However, translating to SAT can cause a significant size increase, restricting its use to smaller instances. To mitigate this, frameworks using multiple local SAT calls for gradually improving a heuristic solution have been proposed. The performance of such algorithmic frameworks heavily relies on critical parameters, including the size of selected local instances and the time allocated per SAT call. This paper examines the automated configuration of the treewidth SAT-based local improvement method (TW-SLIM) framework, which uses multiple SAT calls for computing tree decompositions of small width, a fundamental problem in combinatorial optimization. We explore various TW-SLIM configuration methods, including offline learning and real-time adjustments, significantly outperforming default settings in multi-SAT scenarios with changing problems. Building upon insights gained from offline training and real-time configurations for TW-SLIM, we propose the iterative cascading policy—a novel hybrid technique that uniquely combines both. The iterative cascading policy employs a pool of 30 configurations obtained through clustering-based offline methods, deploying them in dynamic cascades across multiple rounds. In each round, the 30 configurations are tested according to the cascading ordering, and the best tree decomposition is retained for further improvement, with the option to adjust the following ordering of cascades. This iterative approach significantly enhances the performance of TW-SLIM beyond baseline results, even within varying global timeouts. This highlights the effectiveness of the proposed iterative cascading policy in enhancing the efficiency and efficacy of complex algorithmic frameworks like TW-SLIM.

This paper proposes SAT-based techniques to calculate a specific normal form of a given finite mathematical structure (model). The normal form is obtained by permuting the domain elements so that the representation of the structure is lexicographically smallest possible. Such a normal form is of interest to mathematicians as it enables easy cataloging of algebraic structures. In particular, two structures are isomorphic precisely when their normal forms are the same. This form is also natural to inspect as mathematicians have been using it routinely for many decades. We develop a novel approach where a SAT solver is used in a black-box fashion to compute the smallest representative. The approach constructs the representative gradually and searches the space of possible isomorphisms, requiring a small number of variables. However, the approach may lead to a large number of SAT calls and therefore we devise propagation techniques to reduce this number. The paper focuses on finite structures with a single binary operation (encompassing groups, semigroups, etc.). However, the approach is generalizable to arbitrary finite structures. We provide an implementation of the proposed algorithm and evaluate it on a variety of algebraic structures.

Graph matching is a fundamental problem in pattern recognition, with many applications such as software analysis and computational biology. One well-known type of graph matching problem is graph isomorphism, which consists of deciding if two graphs are identical. Despite its usefulness, the properties that one may check using graph isomorphism are rather limited, since it only allows strict equality checks between two graphs. For example, it does not allow one to check complex structural properties such as if the target graph is an arbitrary length sequence followed by an arbitrary size loop. We propose a generalization of graph isomorphism that allows one to check such properties through a declarative specification. This specification is given in the form of a Regular Graph Pattern (ReGaP), a special type of graph, inspired by regular expressions, that may contain wildcard nodes that represent arbitrary structures such as variable-sized sequences or subgraphs. We propose a SAT-based algorithm for checking if a target graph matches a given ReGaP. We also propose a preprocessing technique for improving the performance of the algorithm and evaluate it through an extensive experimental evaluation on benchmarks from the CodeSearchNet dataset.

Optical Flow Estimation aims to find the 2D dense motion field between two frames. Due to the limitation of model structures and training datasets, existing methods often rely too much on local clues and ignore the integrity of objects, resulting in fragmented motion estimation. Through theoretical analysis, we find the pre-trained large vision models are helpful in optical flow estimation, and we notice that the recently famous Segment Anything Model (SAM) demonstrates a strong ability to segment complete objects, which is suitable for solving the fragmentation problem. We thus propose a solution to embed the frozen SAM image encoder into FlowFormer to enhance object perception. To address the challenge of in-depth utilizing SAM in non-segmentation tasks like optical flow estimation, we propose an Optical Flow Task-Specific Adaption scheme, including a Context Fusion Module to fuse the SAM encoder with the optical flow context encoder, and a Context Adaption Module to adapt the SAM features for optical flow task with Learned Task-Specific Embedding. Our proposed SAMFlow model reaches 0.86/2.10 clean/final EPE and 3.55/12.32 EPE/F1-all on Sintel and KITTI-15 training set, surpassing Flowformer by 8.5%/9.9% and 13.2%/16.3%. Furthermore, our model achieves state-of-the-art performance on the Sintel and KITTI-15 benchmarks, ranking #1 among all two-frame methods on Sintel clean pass.

While deep learning models have shown significant performance across various domains, their deployment needs extensive resources and advanced computing infrastructure. As a solution, Machine Learning as a Service (MLaaS) has emerged, lowering the barriers for users to release or productize their deep learning models. However, previous studies have highlighted potential privacy and security concerns associated with MLaaS, and one primary threat is model extraction attacks. To address this, there are many defense solutions but they suffer from unrealistic assumptions and generalization issues, making them less practical for reliable protection. Driven by these limitations, we introduce a novel defense mechanism, SAME, based on the concept of sample reconstruction. This strategy imposes minimal prerequisites on the defender's capabilities, eliminating the need for auxiliary Out-of-Distribution (OOD) datasets, user query history, white-box model access, and additional intervention during model training. It is compatible with existing active defense methods. Our extensive experiments corroborate the superior efficacy of SAME over state-of-the-art solutions. Our code is available at https://github.com/xythink/SAME.