اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدThere is More than Meets the Eye: Self-Supervised Multi-Object Detection and Tracking with Sound by Distilling Multimodal Knowledge
206
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Attributes of sound inherent to objects can provide valuable cues to learn rich representations for object detection and tracking. Furthermore, the co-occurrence of audiovisual events in videos can be exploited to localize objects over the image field by solely monitoring the sound in the environment. Thus far, this has only been feasible in scenarios where the camera is static and for single object detection. Moreover, the robustness of these methods has been limited as they primarily rely on RGB images which are highly susceptible to illumination and weather changes. In this work, we present the novel self-supervised MM-DistillNet framework consisting of multiple teachers that leverage diverse modalities including RGB, depth and thermal images, to simultaneously exploit complementary cues and distill knowledge into a single audio student network. We propose the new MTA loss function that facilitates the distillation of information from multimodal teachers in a self-supervised manner. Additionally, we propose a novel self-supervised pretext task for the audio student that enables us to not rely on labor-intensive manual annotations. We introduce a large-scale multimodal dataset with over 113,000 time-synchronized frames of RGB, depth, thermal, and audio modalities. Extensive experiments demonstrate that our approach outperforms state-of-the-art methods while being able to detect multiple objects using only sound during inference and even while moving.
قيم البحث
اقرأ أيضاً
The presence of submicron grains has been inferred in several debris discs, despite the fact that these particles should be blown out by stellar radiation pressure on very short timescales. So far, no fully satisfying explanation has been found for t
his apparent paradox. We investigate the possibility that the observed abundances of submicron grains could be naturally produced in bright debris discs, where the high collisional activity produces them at a rate high enough to partially compensate for their rapid removal. We also investigate to what extent this potential presence of small grains can affect our understanding of some debris disc characteristics. We use a code following the collisional evolution of a debris disc down to submicron grains far below the limiting blow-out size $s_{blow}$. We explore different configurations: A and G stars, cold and warm discs, bright and very bright systems. We find that, in bright discs (fractional luminosity $>10^{-3}$) around A stars, there is always a high-enough amount of submicron grains to leave detectable signatures, both in scattered-light, where the discs color becomes blue, and in the mid-IR ($10<lambda<20mu$m), where it boosts the discs luminosity by at least a factor of 2 and induces a pronounced silicate solid-state band around $10mu$m. We also show that, with this additional contribution of submicron grains, the SED can mimic that of two debris belts separated by a factor of 2 in radial distance. For G stars, the effect of $s<s_{blow}$ grains remains limited in the spectra, in spite of the fact that they dominate the systems geometrical cross section. We also find that, for all considered cases, the halo of small (bound and unbound) grains that extends far beyond the main disc contributes to $sim50$% of the flux up to $lambdasim50mu$m wavelengths.
Humans are able to localize objects in the environment using both visual and auditory cues, integrating information from multiple modalities into a common reference frame. We introduce a system that can leverage unlabeled audio-visual data to learn t
o localize objects (moving vehicles) in a visual reference frame, purely using stereo sound at inference time. Since it is labor-intensive to manually annotate the correspondences between audio and object bounding boxes, we achieve this goal by using the co-occurrence of visual and audio streams in unlabeled videos as a form of self-supervision, without resorting to the collection of ground-truth annotations. In particular, we propose a framework that consists of a vision teacher network and a stereo-sound student network. During training, knowledge embodied in a well-established visual vehicle detection model is transferred to the audio domain using unlabeled videos as a bridge. At test time, the stereo-sound student network can work independently to perform object localization us-ing just stereo audio and camera meta-data, without any visual input. Experimental results on a newly collected Au-ditory Vehicle Tracking dataset verify that our proposed approach outperforms several baseline approaches. We also demonstrate that our cross-modal auditory localization approach can assist in the visual localization of moving vehicles under poor lighting conditions.
In this paper, we challenge the common assumption that collapsing the spatial dimensions of a 3D (spatial-channel) tensor in a convolutional neural network (CNN) into a vector via global pooling removes all spatial information. Specifically, we demon
strate that positional information is encoded based on the ordering of the channel dimensions, while semantic information is largely not. Following this demonstration, we show the real world impact of these findings by applying them to two applications. First, we propose a simple yet effective data augmentation strategy and loss function which improves the translation invariance of a CNNs output. Second, we propose a method to efficiently determine which channels in the latent representation are responsible for (i) encoding overall position information or (ii) region-specific positions. We first show that semantic segmentation has a significant reliance on the overall position channels to make predictions. We then show for the first time that it is possible to perform a `region-specific attack, and degrade a networks performance in a particular part of the input. We believe our findings and demonstrated applications will benefit research areas concerned with understanding the characteristics of CNNs.
Comprehensive understanding of dynamic scenes is a critical prerequisite for intelligent robots to autonomously operate in their environment. Research in this domain, which encompasses diverse perception problems, has primarily been focused on addres
sing specific tasks individually rather than modeling the ability to understand dynamic scenes holistically. In this paper, we introduce a novel perception task denoted as multi-object panoptic tracking (MOPT), which unifies the conventionally disjoint tasks of semantic segmentation, instance segmentation, and multi-object tracking. MOPT allows for exploiting pixel-level semantic information of thing and stuff classes, temporal coherence, and pixel-level associations over time, for the mutual benefit of each of the individual sub-problems. To facilitate quantitative evaluations of MOPT in a unified manner, we propose the soft panoptic tracking quality (sPTQ) metric. As a first step towards addressing this task, we propose the novel PanopticTrackNet architecture that builds upon the state-of-the-art top-down panoptic segmentation network EfficientPS by adding a new tracking head to simultaneously learn all sub-tasks in an end-to-end manner. Additionally, we present several strong baselines that combine predictions from state-of-the-art panoptic segmentation and multi-object tracking models for comparison. We present extensive quantitative and qualitative evaluations of both vision-based and LiDAR-based MOPT that demonstrate encouraging results.
In this paper, we propose an effective knowledge transfer framework to boost the weakly supervised object detection accuracy with the help of an external fully-annotated source dataset, whose categories may not overlap with the target domain. This se
tting is of great practical value due to the existence of many off-the-shelf detection datasets. To more effectively utilize the source dataset, we propose to iteratively transfer the knowledge from the source domain by a one-class universal detector and learn the target-domain detector. The box-level pseudo ground truths mined by the target-domain detector in each iteration effectively improve the one-class universal detector. Therefore, the knowledge in the source dataset is more thoroughly exploited and leveraged. Extensive experiments are conducted with Pascal VOC 2007 as the target weakly-annotated dataset and COCO/ImageNet as the source fully-annotated dataset. With the proposed solution, we achieved an mAP of $59.7%$ detection performance on the VOC test set and an mAP of $60.2%$ after retraining a fully supervised Faster RCNN with the mined pseudo ground truths. This is significantly better than any previously known results in related literature and sets a new state-of-the-art of weakly supervised object detection under the knowledge transfer setting. Code: url{https://github.com/mikuhatsune/wsod_transfer}.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
