United States

Domain generalization in semantic segmentation aims to alleviate the performance degradation on unseen domains through learning domain-invariant features. Existing methods diversify images in the source domain by adding complex or even abnormal textures to reduce the sensitivity to domain-specific features. However, these approaches depends heavily on the richness of the texture bank and training them can be time-consuming. In contrast to importing textures arbitrarily or augmenting styles randomly, we focus on the single source domain itself to achieve the generalization. In this paper, we present a novel adaptive texture filtering mechanism to suppress the influence of texture without using augmentation, thus eliminating the interference of domain-specific features. Further, we design a hierarchical guidance generalization network equipped with structure-guided enhancement modules, which purpose to learn the domain-invariant generalized knowledge. Extensive experiments together with ablation studies on widely-used datasets are conducted to verify the effectiveness of the proposed model, and reveal its superiority over other state-of-the-art alternatives. 

AAAI 2023

Adaptive Texture Filtering for Single-domain Generalized Segmentation

representation learning for vision

segmentation

computer vision

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

AAAI 2022

poster

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-23 is the Thirty-Seventh AAAI Conference on Artificial Intelligence. The theme of this conference is to create collaborative bridges within and beyond AI. Like previous AAAI conferences, AAAI-23 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and two new activities: a Bridge Program and a Lab Program. Many of these activities are tailored to the theme of bridges and all are selected according to the highest standards, with additional programs for students and young researchers. 
AAAI is providing you with a conference planner, which you can use to help organize your itinerary of activities. This includes talks to attend in person, talks to attend remotely, breaks with colleagues and your site seeing activities. To access this conference planner, please go to [https://aaai-2023.takemobi.io](https://aaai-2023.takemobi.io).

In order to access this site, you need to register. If you haven't already, please register [here](https://aaai.org/Conferences/AAAI-23/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 for the first time explores audio-visual event localization in an unsupervised manner. Previous methods tackle this problem in a supervised setting and require segment-level or video-level event category ground-truth to train the model. However, building large-scale multi-modality datasets with category annotations is human-intensive and thus not scalable to real-world applications. To this end, we propose cross-modal label contrastive learning to exploit multi-modal information among unlabeled audio and visual streams as self-supervision signals. At the feature representation level, multi-modal representations are collaboratively learned from audio and visual components by using self-supervised representation learning. At the label level, we propose a novel self-supervised pretext task i.e. label contrasting to self-annotate videos with pseudo-labels for localization model training. Note that irrelevant background would hinder the acquisition of high-quality pseudo-labels and thus lead to an inferior localization model. To address this issue, we then propose an expectation-maximization algorithm that optimizes the pseudo-label acquisition and localization model in a coarse-to-fine manner. Extensive experiments demonstrate that our unsupervised approach performs reasonably well compared to the state-of-the-art supervised methods.

Cross-Modal Label Contrastive Learning for Unsupervised Audio-Visual Event Localization

Most of the existing incomplete multi-view clustering (IMVC) methods focus on attaining a consensus representation from different views but ignore the important information hidden in the missing views and the latent intrinsic structures in each view. To tackle these issues, in this paper, a unified and novel framework, named tensorized incomplete multi-view clustering with intrinsic graph completion (TIMVC_IGC) is proposed. Firstly, owing to the effectiveness of the low-rank representation in revealing the inherent structure of the data, we exploit it to infer the missing instances and construct the complete graph for each view. Afterwards, inspired by the structural consistency, a between-view consistency constraint is imposed to guarantee the similarity of the graphs from different views. More importantly, the TIMVC_IGC simultaneously learns the low-rank structures of the different views and explores the correlations of the different graphs in a latent manifold sub-space using a low-rank tensor constraint, such that the intrinsic graphs of the different views can be obtained. Finally, a consensus representation for each sample is gained with a co-regularization term for final clustering.  Experimental results on several real-world databases illustrates that the proposed method can outperform the other state-of-the-art related methods for incomplete multi-view clustering.

Tensorized Incomplete Multiview Clustering with Intrinsic Graph Completion

Mixup provides interpolated training samples and allows the model to obtain smoother decision boundaries for better generalization. The idea can be naturally applied to the domain adaptation task, where we can mix the source and target samples to obtain domain-mixed samples for better adaptation. However, the extension of the idea from classification to segmentation (i.e., structured output) is nontrivial. This paper systematically studies the impact of mixup under the domain adaptive semantic segmentation task and presents a simple yet effective mixup strategy called Bidirectional Domain Mixup (BDM). In specific, we achieve domain mixup in two-step: cut and paste. Given the warm-up model trained from any adaptation techniques, we forward the source and target samples and perform a simple threshold-based cut out of the unconfident regions (cut). After then, we fill-in the dropped regions with the other domain region patches (paste). In doing so, we jointly consider class distribution, spatial structure, and pseudo label confidence. Based on our analysis, we found that BDM leaves domain transferable regions by cutting, balances the dataset-level class distribution while preserving natural scene context by pasting. We coupled our proposal with various state-of-the-art adaptation models and observe significant improvement consistently. We also provide extensive ablation experiments to empirically verify our main components of the framework. Visit our project page with the code at https://sites.google.com/view/bidirectional-domain-mixup

Bidirectional Domain Mixup for Domain Adaptive Semantic Segmentation

A plethora of previous studies indicates that making full use of multifarious intrinsic properties of primordial data is a valid pathway to recover original images from their degraded observations. Typically, both low-rankness and local-smoothness broadly exist in real-world tensor data such as hyperspectral images and videos. Modeling based on both properties has received a great deal of attention, whereas most studies concentrate on experimental performance, and theoretical investigations are still lacking. In this paper, we study the tensor compressive sensing problem based on the correlated total variation norm, which is a new regularizer used to simultaneously capture both properties existing in the same data set. The new regularizer has the outstanding advantage of not using a trade-off parameter to balance the two properties. The obtained theories provide a robust recovery guarantee, where the error bound shows that our model certainly benefits from both properties in ground-truth data adaptively. Moreover, based on the ADMM update procedure, we design an algorithm with a global convergence guarantee to solve this model. At last, we carry out experiments to apply our model to hyperspectral image and video restoration problems. The experimental results show that our method is prominently better than many other competing ones. In particular, for the hyperspectral image restoration task, our method achieves satisfactory results under an extremely low sampling ratio 1%.

Tensor Compressive Sensing Fused Low-rankness and Local-smoothness

We investigate the problem of distributed online convex optimization with complicated constraints, in which the projection operation could be the computational bottleneck. To avoid projections, distributed online projection-free methods have been proposed and attain an $\mathcal{O}(T^{3/4})$ regret bound for general convex losses. However, they cannot utilize the smoothness condition, which has been exploited in the centralized setting to improve the regret. In this paper, we propose a new distributed online projection-free method with a tighter regret bound of $\mathcal{O}(T^{2/3})$ for smooth and convex losses. Specifically, we first provide a distributed extension of Follow-the-Perturbed-Leader so that the smoothness can be utilized in the distributed setting. Then, we reduce the computational cost via sampling and blocking techniques. In this way, our method only needs to solve one linear optimization per round on average. Finally, we conduct experiments on benchmark datasets to verify the effectiveness of our proposed method. 

Distributed Projection-free Online Learning for Smooth and Convex Losses

RGB-D object tracking has attracted considerable attention recently, achieving promising performance thanks to the symbiosis between visual and depth channels. However, given a limited amount of annotated RGB-D tracking data, most state-of-the-art RGB-D trackers are simple extensions of high-performance RGB-only trackers, without fully exploiting the underlying potential of the depth channel in the offline training stage. To address the dataset deficiency issue, a new RGB-D dataset named RGBD1K is released in this paper. The RGBD1K contains 1,050 sequences with about 2.5M frames in total. To demonstrate the benefits of training on a larger RGB-D data set in general, and RGBD1K in particular, we develop a transformer-based RGB-D tracker, named SPT, as a baseline for future visual object tracking studies using the new dataset. The results, of extensive experiments using the SPT tracker demonstrate the potential of the RGBD1K dataset to improve the performance of RGB-D tracking, inspiring future developments of effective tracker designs. The dataset and codes will be available on the project homepage: https://github.com/xuefeng-zhu5/RGBD1K. 

RGBD1K: A Large-scale Dataset and Benchmark for RGB-D Object Tracking

Vision-Language Navigation (VLN) is a challenging task which requires an agent to align complex visual observations to language instructions to reach the goal position. Most existing VLN agents directly learn to align the raw directional features and visual features trained using one-hot labels to linguistic instruction features. However, the big semantic gap among these multi-modal inputs makes the alignment difficult and therefore limits the navigation performance. In this paper, we propose Actional Atomic-Concept Learning (AACL), which maps visual observations to actional atomic concepts for facilitating the alignment. Specifically, an actional atomic concept is a natural language phrase containing an atomic action and an object, e.g., ``go up stairs''. These actional atomic concepts, which serve as the bridge between observations and instructions, can effectively mitigate the semantic gap and simplify the alignment. AACL contains three core components: 1) a concept mapping module to map the observations to the actional atomic concept representations through the VLN environment and the recently proposed Contrastive Language-Image Pretraining (CLIP) model, 2) a concept refining adapter to encourage more instruction-oriented object concept extraction by re-ranking the predicted object concepts by CLIP, and 3) an observation co-embedding module which utilizes concept representations to regularize the observation representations. Our AACL establishes new state-of-the-art results on both fine-grained (R2R) and high-level (REVERIE and R2R-Last) VLN benchmarks. Moreover, the visualization shows that AACL significantly improves the interpretability in action decision. Code  will be available at https://gitee.com/mindspore/models/tree/master/research/cv/VLN-AACL.

Actional Atomic-Concept Learning for Demystifying Vision-Language Navigation

Dense passage retrieval aims to retrieve the relevant passages of a query from a large amount of corpus based on pre-trained language models. Recent studies have explored improving pre-trained language models to boost dense retrieval performance. This paper proposes CoT-MAE (ConTextual Masked Auto-Encoder), a simple yet effective generative pre-training method for dense passage retrieval. CoT-MAE employs an asymmetric encoder-decoder architecture that learns to compress the sentence semantics into a dense vector through self-supervised and context-supervised masked auto-encoding. Precisely, self-supervised masked auto-encoding learns to model the semantics of the tokens inside a text span, and context-supervised masked auto-encoding learns to model the semantical correlation between the text spans. We conduct experiments on large-scale passage retrieval benchmarks and show substantial improvements over strong baselines, demonstrating the high efficiency of CoT-MAE. Moreover, CoT-MAE is based on the most commonly used masked auto-encoding task, only requires access to unlabeled data, and is broadly applicable in other domains.

ConTextual Mask AutoEncoder for Dense Passage Retrieval

Weakly-supervised semantic segmentation aims to train a semantic segmentation network using weak labels. Among weak labels, image-level label has been the most popular choice due to its simplicity. However, since image-level labels lack accurate object region information, additional modules such as saliency detector have been exploited in weakly supervised semantic segmentation, which requires pixel-level label for training. In this paper, we explore a self-supervised vision transformer to mitigate the heavy efforts on generation of pixel-level annotations. By exploiting the features obtained from self-supervised vision transformer, our superpixel discovery method finds out the semantic-aware superpixels based on the feature similarity in an unsupervised manner. Once we obtain the superpixels, we train the semantic segmentation network using superpixel-guided seeded region growing method. Despite its simplicity, our approach achieves the competitive result with the state-of-the-arts on PASCAL VOC 2012 and MS-COCO 2014 semantic segmentation datasets for weakly supervised semantic segmentation.

Downloads

Next from AAAI 2023

Cross-Modal Label Contrastive Learning for Unsupervised Audio-Visual Event Localization

Similar lecture

DIRL: Domain-Invariant Representation Learning for Generalizable Semantic Segmentation