Visual-audio navigation (VAN) is attracting more and more attention from the robotic community due to its broad applications, \emph{e.g.}, household robots and rescue robots. In this task, an embodied agent must search for and navigate to the sound source with egocentric visual and audio observations. However, the existing methods are limited in two aspects: 1) poor generalization to unheard sound categories; 2) sample inefficient in training. Focusing on these two problems, we propose a brain-inspired plug-and-play method to learn a semantic-agnostic and spatial-aware representation for generalizable visual-audio navigation. We meticulously design two auxiliary tasks for respectively accelerating learning representations with the above-desired characteristics. With these two auxiliary tasks, the agent learns a spatially-correlated representation of visual and audio inputs that can be applied to work on environments with novel sounds and maps. Experiment results on realistic 3D scenes (Replica and Matterport3D) demonstrate that our method achieves better generalization performance when zero-shot transferred to scenes with unseen maps and unheard sound categories.
相關內容
Zero-shot video recognition (ZSVR) is a task that aims to recognize video categories that have not been seen during the model training process. Recently, vision-language models (VLMs) pre-trained on large-scale image-text pairs have demonstrated impressive transferability for ZSVR. To make VLMs applicable to the video domain, existing methods often use an additional temporal learning module after the image-level encoder to learn the temporal relationships among video frames. Unfortunately, for video from unseen categories, we observe an abnormal phenomenon where the model that uses spatial-temporal feature performs much worse than the model that removes temporal learning module and uses only spatial feature. We conjecture that improper temporal modeling on video disrupts the spatial feature of the video. To verify our hypothesis, we propose Feature Factorization to retain the orthogonal temporal feature of the video and use interpolation to construct refined spatial-temporal feature. The model using appropriately refined spatial-temporal feature performs better than the one using only spatial feature, which verifies the effectiveness of the orthogonal temporal feature for the ZSVR task. Therefore, an Orthogonal Temporal Interpolation module is designed to learn a better refined spatial-temporal video feature during training. Additionally, a Matching Loss is introduced to improve the quality of the orthogonal temporal feature. We propose a model called OTI for ZSVR by employing orthogonal temporal interpolation and the matching loss based on VLMs. The ZSVR accuracies on popular video datasets (i.e., Kinetics-600, UCF101 and HMDB51) show that OTI outperforms the previous state-of-the-art method by a clear margin.
When labeled data is insufficient, semi-supervised learning with the pseudo-labeling technique can significantly improve the performance of automatic speech recognition. However, pseudo-labels are often noisy, containing numerous incorrect tokens. Taking noisy labels as ground-truth in the loss function results in suboptimal performance. Previous works attempted to mitigate this issue by either filtering out the nosiest pseudo-labels or improving the overall quality of pseudo-labels. While these methods are effective to some extent, it is unrealistic to entirely eliminate incorrect tokens in pseudo-labels. In this work, we propose a novel framework named alternative pseudo-labeling to tackle the issue of noisy pseudo-labels from the perspective of the training objective. The framework comprises several components. Firstly, a generalized CTC loss function is introduced to handle noisy pseudo-labels by accepting alternative tokens in the positions of incorrect tokens. Applying this loss function in pseudo-labeling requires detecting incorrect tokens in the predicted pseudo-labels. In this work, we adopt a confidence-based error detection method that identifies the incorrect tokens by comparing their confidence scores with a given threshold, thus necessitating the confidence score to be discriminative. Hence, the second proposed technique is the contrastive CTC loss function that widens the confidence gap between the correctly and incorrectly predicted tokens, thereby improving the error detection ability. Additionally, obtaining satisfactory performance with confidence-based error detection typically requires extensive threshold tuning. Instead, we propose an automatic thresholding method that uses labeled data as a proxy for determining the threshold, thus saving the pain of manual tuning.
Edge-device co-inference refers to deploying well-trained artificial intelligent (AI) models at the network edge under the cooperation of devices and edge servers for providing ambient intelligent services. For enhancing the utilization of limited network resources in edge-device co-inference tasks from a systematic view, we propose a task-oriented scheme of integrated sensing, computation and communication (ISCC) in this work. In this system, all devices sense a target from the same wide view to obtain homogeneous noise-corrupted sensory data, from which the local feature vectors are extracted. All local feature vectors are aggregated at the server using over-the-air computation (AirComp) in a broadband channel with the orthogonal-frequency-division-multiplexing technique for suppressing the sensing and channel noise. The aggregated denoised global feature vector is further input to a server-side AI model for completing the downstream inference task. A novel task-oriented design criterion, called maximum minimum pair-wise discriminant gain, is adopted for classification tasks. It extends the distance of the closest class pair in the feature space, leading to a balanced and enhanced inference accuracy. Under this criterion, a problem of joint sensing power assignment, transmit precoding and receive beamforming is formulated. The challenge lies in three aspects: the coupling between sensing and AirComp, the joint optimization of all feature dimensions' AirComp aggregation over a broadband channel, and the complicated form of the maximum minimum pair-wise discriminant gain. To solve this problem, a task-oriented ISCC scheme with AirComp is proposed. Experiments based on a human motion recognition task are conducted to verify the advantages of the proposed scheme over the existing scheme and a baseline.
Click-Through Rate (CTR) prediction, estimating the probability of a user clicking on an item, is essential in industrial applications, such as online advertising. Many works focus on user behavior modeling to improve CTR prediction performance. However, most of those methods only model users' positive interests from users' click items while ignoring the context information, which is the display items around the clicks, resulting in inferior performance. In this paper, we highlight the importance of context information on user behavior modeling and propose a novel model named Deep Context Interest Network (DCIN), which integrally models the click and its display context to learn users' context-aware interests. DCIN consists of three key modules: 1) Position-aware Context Aggregation Module (PCAM), which performs aggregation of display items with an attention mechanism; 2) Feedback-Context Fusion Module (FCFM), which fuses the representation of clicks and display contexts through non-linear feature interaction; 3) Interest Matching Module (IMM), which activates interests related with the target item. Moreover, we provide our hands-on solution to implement our DCIN model on large-scale industrial systems. The significant improvements in both offline and online evaluations demonstrate the superiority of our proposed DCIN method. Notably, DCIN has been deployed on our online advertising system serving the main traffic, which brings 1.5% CTR and 1.5% RPM lift.
Multi-object tracking (MOT) at low frame rates can reduce computational, storage and power overhead to better meet the constraints of edge devices. Many existing MOT methods suffer from significant performance degradation in low-frame-rate videos due to significant location and appearance changes between adjacent frames. To this end, we propose to explore collaborative tracking learning (ColTrack) for frame-rate-insensitive MOT in a query-based end-to-end manner. Multiple historical queries of the same target jointly track it with richer temporal descriptions. Meanwhile, we insert an information refinement module between every two temporal blocking decoders to better fuse temporal clues and refine features. Moreover, a tracking object consistency loss is proposed to guide the interaction between historical queries. Extensive experimental results demonstrate that in high-frame-rate videos, ColTrack obtains higher performance than state-of-the-art methods on large-scale datasets Dancetrack and BDD100K, and outperforms the existing end-to-end methods on MOT17. More importantly, ColTrack has a significant advantage over state-of-the-art methods in low-frame-rate videos, which allows it to obtain faster processing speeds by reducing frame-rate requirements while maintaining higher performance. Code will be released at //github.com/yolomax/ColTrack
Machine learning (ML) is widely used today, especially through deep neural networks (DNNs), however, increasing computational load and resource requirements have led to cloud-based solutions. To address this problem, a new generation of networks called Spiking Neural Networks (SNN) has emerged, which mimic the behavior of the human brain to improve efficiency and reduce energy consumption. These networks often process large amounts of sensitive information, such as confidential data, and thus privacy issues arise. Homomorphic encryption (HE) offers a solution, allowing calculations to be performed on encrypted data without decrypting it. This research compares traditional DNNs and SNNs using the Brakerski/Fan-Vercauteren (BFV) encryption scheme. The LeNet-5 model, a widely-used convolutional architecture, is used for both DNN and SNN models based on the LeNet-5 architecture, and the networks are trained and compared using the FashionMNIST dataset. The results show that SNNs using HE achieve up to 40% higher accuracy than DNNs for low values of the plaintext modulus t, although their execution time is longer due to their time-coding nature with multiple time-steps.
Visual information extraction (VIE) has attracted considerable attention recently owing to its various advanced applications such as document understanding, automatic marking and intelligent education. Most existing works decoupled this problem into several independent sub-tasks of text spotting (text detection and recognition) and information extraction, which completely ignored the high correlation among them during optimization. In this paper, we propose a robust visual information extraction system (VIES) towards real-world scenarios, which is a unified end-to-end trainable framework for simultaneous text detection, recognition and information extraction by taking a single document image as input and outputting the structured information. Specifically, the information extraction branch collects abundant visual and semantic representations from text spotting for multimodal feature fusion and conversely, provides higher-level semantic clues to contribute to the optimization of text spotting. Moreover, regarding the shortage of public benchmarks, we construct a fully-annotated dataset called EPHOIE (//github.com/HCIILAB/EPHOIE), which is the first Chinese benchmark for both text spotting and visual information extraction. EPHOIE consists of 1,494 images of examination paper head with complex layouts and background, including a total of 15,771 Chinese handwritten or printed text instances. Compared with the state-of-the-art methods, our VIES shows significant superior performance on the EPHOIE dataset and achieves a 9.01% F-score gain on the widely used SROIE dataset under the end-to-end scenario.
Learning latent representations of nodes in graphs is an important and ubiquitous task with widespread applications such as link prediction, node classification, and graph visualization. Previous methods on graph representation learning mainly focus on static graphs, however, many real-world graphs are dynamic and evolve over time. In this paper, we present Dynamic Self-Attention Network (DySAT), a novel neural architecture that operates on dynamic graphs and learns node representations that capture both structural properties and temporal evolutionary patterns. Specifically, DySAT computes node representations by jointly employing self-attention layers along two dimensions: structural neighborhood and temporal dynamics. We conduct link prediction experiments on two classes of graphs: communication networks and bipartite rating networks. Our experimental results show that DySAT has a significant performance gain over several different state-of-the-art graph embedding baselines.
We consider the problem of referring image segmentation. Given an input image and a natural language expression, the goal is to segment the object referred by the language expression in the image. Existing works in this area treat the language expression and the input image separately in their representations. They do not sufficiently capture long-range correlations between these two modalities. In this paper, we propose a cross-modal self-attention (CMSA) module that effectively captures the long-range dependencies between linguistic and visual features. Our model can adaptively focus on informative words in the referring expression and important regions in the input image. In addition, we propose a gated multi-level fusion module to selectively integrate self-attentive cross-modal features corresponding to different levels in the image. This module controls the information flow of features at different levels. We validate the proposed approach on four evaluation datasets. Our proposed approach consistently outperforms existing state-of-the-art methods.
Top-down visual attention mechanisms have been used extensively in image captioning and visual question answering (VQA) to enable deeper image understanding through fine-grained analysis and even multiple steps of reasoning. In this work, we propose a combined bottom-up and top-down attention mechanism that enables attention to be calculated at the level of objects and other salient image regions. This is the natural basis for attention to be considered. Within our approach, the bottom-up mechanism (based on Faster R-CNN) proposes image regions, each with an associated feature vector, while the top-down mechanism determines feature weightings. Applying this approach to image captioning, our results on the MSCOCO test server establish a new state-of-the-art for the task, achieving CIDEr / SPICE / BLEU-4 scores of 117.9, 21.5 and 36.9, respectively. Demonstrating the broad applicability of the method, applying the same approach to VQA we obtain first place in the 2017 VQA Challenge.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191