NeurIPS 2023 论文列表

Audio-Visual Source Localization (AVSL) aims to locate sounding objects within video frames given the paired audio clips. Existing methods predominantly rely on self-supervised contrastive learning of audio-visual correspondence. Without any bounding-box annotations, they struggle to achieve precise localization, especially for small objects, and suffer from blurry boundaries and false positives. Moreover, the naive semi-supervised method is poor in effectively utilizing the abundance of unlabeled audio-visual pairs. In this paper, we propose a novel Semi-Supervised Learning framework for AVSL, namely Dual Mean-Teacher (DMT), comprising two teacher-student structures to circumvent the confirmation bias issue. Specifically, two teachers, pre-trained on limited labeled data, are employed to filter out noisy samples via the consensus between their predictions, and then generate high-quality pseudo-labels by intersecting their confidence maps. The optimal utilization of both labeled and unlabeled data combined with this unbiased framework enable DMT to outperform current state-of-the-art methods by a large margin, with CIoU of $\textbf{90.4\%}$ and $\textbf{48.8\%}$ on Flickr-SoundNet and VGG-Sound Source, obtaining $\textbf{8.9\%}$ and $\textbf{9.6\%}$ improvements respectively, given only $3\%$ of data positional-annotated. We also extend our framework to some existing AVSL methods and consistently boost their performance. Our code is publicly available at https://github.com/gyx-gloria/DMT.

In recent years, the deployment of large-scale pre-trained models in audio-visual downstream tasks has yielded remarkable outcomes. However, these models, primarily trained on single-modality unconstrained datasets, still encounter challenges in feature extraction for multi-modal tasks, leading to suboptimal performance. This limitation arises due to the introduction of irrelevant modality-specific information during encoding, which adversely affects the performance of downstream tasks. To address this challenge, this paper proposes a novel Dual-Guided Spatial-Channel-Temporal (DG-SCT) attention mechanism. This mechanism leverages audio and visual modalities as soft prompts to dynamically adjust the parameters of pre-trained models based on the current multi-modal input features. Specifically, the DG-SCT module incorporates trainable cross-modal interaction layers into pre-trained audio-visual encoders, allowing adaptive extraction of crucial information from the current modality across spatial, channel, and temporal dimensions, while preserving the frozen parameters of large-scale pre-trained models. Experimental evaluations demonstrate that our proposed model achieves state-of-the-art results across multiple downstream tasks, including AVE, AVVP, AVS, and AVQA. Furthermore, our model exhibits promising performance in challenging few-shot and zero-shot scenarios. The source code and pre-trained models are available at https://github.com/haoyi-duan/DG-SCT.

Human visual reasoning is characterized by an ability to identify abstract patterns from only a small number of examples, and to systematically generalize those patterns to novel inputs. This capacity depends in large part on our ability to represent complex visual inputs in terms of both objects and relations. Recent work in computer vision has introduced models with the capacity to extract object-centric representations, leading to the ability to process multi-object visual inputs, but falling short of the systematic generalization displayed by human reasoning. Other recent models have employed inductive biases for relational abstraction to achieve systematic generalization of learned abstract rules, but have generally assumed the presence of object-focused inputs. Here, we combine these two approaches, introducing Object-Centric Relational Abstraction (OCRA), a model that extracts explicit representations of both objects and abstract relations, and achieves strong systematic generalization in tasks (including a novel dataset, CLEVR-ART, with greater visual complexity) involving complex visual displays.

This paper studies the fundamental learning problem of the energy-based model (EBM). Learning the EBM can be achieved using the maximum likelihood estimation (MLE), which typically involves the Markov Chain Monte Carlo (MCMC) sampling, such as the Langevin dynamics. However, the noise-initialized Langevin dynamics can be challenging in practice and hard to mix. This motivates the exploration of joint training with the generator model where the generator model serves as a complementary model to bypass MCMC sampling. However, such a method can be less accurate than the MCMC and result in biased EBM learning. While the generator can also serve as an initializer model for better MCMC sampling, its learning can be biased since it only matches the EBM and has no access to empirical training examples. Such biased generator learning may limit the potential of learning the EBM. To address this issue, we present a joint learning framework that interweaves the maximum likelihood learning algorithm for both the EBM and the complementary generator model. In particular, the generator model is learned by MLE to match both the EBM and the empirical data distribution, making it a more informative initializer for MCMC sampling of EBM. Learning generator with observed examples typically requires inference of the generator posterior. To ensure accurate and efficient inference, we adopt the MCMC posterior sampling and introduce a complementary inference model to initialize such latent MCMC sampling. We show that three separate models can be seamlessly integrated into our joint framework through two (dual-) MCMC teaching, enabling effective and efficient EBM learning.

Heterogeneous federated learning without assuming any structure is challenging due to the conflicts among non-identical data distributions of clients. In practice, clients often comprise near-homogeneous clusters so training a server-side model per cluster mitigates the conflicts. However, FL with client clustering often suffers from “clustering collapse'', i.e., one cluster's model excels on increasing clients, and reduces to single-model FL. Moreover, cluster-wise models hinder knowledge sharing between clusters and each model depends on fewer clients. Furthermore, the static clustering assumption on data may not hold for dynamically changing models, which are sensitive to cluster imbalance/initialization or outliers. To address these challenges, we propose ''Clustered Additive Modeling (CAM)'', which applies a globally shared model $\Theta_g$ on top of the cluster-wise models $\Theta_{1:K}$, i.e., $y=h(x;\Theta_g)+f(x;\Theta_k)$ for clients of cluster-$k$. The global model captures the features shared by all clusters so $\Theta_{1:K}$ are enforced to focus on the difference among clusters. To train CAM, we develop a novel Fed-CAM algorithm that alternates between client clustering and training global/cluster models to predict the residual of each other. We can easily modify any existing clustered FL methods by CAM and significantly improve their performance without ‘’clustering collapse'' in different non-IID settings. We also provide a convergence analysis of Fed-CAM algorithm.

The task of Visual Object Navigation (VON) involves an agent's ability to locate a particular object within a given scene. To successfully accomplish the VON task, two essential conditions must be fulfiled: 1) the user knows the name of the desired object; and 2) the user-specified object actually is present within the scene. To meet these conditions, a simulator can incorporate predefined object names and positions into the metadata of the scene. However, in real-world scenarios, it is often challenging to ensure that these conditions are always met. Humans in an unfamiliar environment may not know which objects are present in the scene, or they may mistakenly specify an object that is not actually present. Nevertheless, despite these challenges, humans may still have a demand for an object, which could potentially be fulfilled by other objects present within the scene in an equivalent manner. Hence, this paper proposes Demand-driven Navigation (DDN), which leverages the user's demand as the task instruction and prompts the agent to find an object which matches the specified demand. DDN aims to relax the stringent conditions of VON by focusing on fulfilling the user's demand rather than relying solely on specified object names. This paper proposes a method of acquiring textual attribute features of objects by extracting common sense knowledge from a large language model (LLM). These textual attribute features are subsequently aligned with visual attribute features using Contrastive Language-Image Pre-training (CLIP). Incorporating the visual attribute features as prior knowledge, enhances the navigation process. Experiments on AI2Thor with the ProcThor dataset demonstrate that the visual attribute features improve the agent's navigation performance and outperform the baseline methods commonly used in the VON and VLN task and methods with LLMs. The codes and demonstrations can be viewed at https://sites.google.com/view/demand-driven-navigation.

Visual Query Localization on long-form egocentric videos requires spatio-temporal search and localization of visually specified objects and is vital to build episodic memory systems. Prior work develops complex multi-stage pipelines that leverage well-established object detection and tracking methods to perform VQL. However, each stage is independently trained and the complexity of the pipeline results in slow inference speeds. We propose VQLoC, a novel single-stage VQL framework that is end-to-end trainable. Our key idea is to first build a holistic understanding of the query-video relationship and then perform spatio-temporal localization in a single shot manner. Specifically, we establish the query-video relationship by jointly considering query-to-frame correspondences between the query and each video frame and frame-to-frame correspondences between nearby video frames. Our experiments demonstrate that our approach outperforms prior VQL methods by $20$% accuracy while obtaining a $10\times$ improvement in inference speed. VQLoC is also the top entry on the Ego4D VQ2D challenge leaderboard.