Autonomously navigating a robot in everyday crowded spaces requires solving complex perception and planning challenges. When using only monocular image sensor data as input, classical two-dimensional planning approaches cannot be used. While images present a significant challenge when it comes to perception and planning, they also allow capturing potentially important details, such as complex geometry, body movement, and other visual cues. In order to successfully solve the navigation task from only images, algorithms must be able to model the scene and its dynamics using only this channel of information. We investigate whether the world model concept, which has shown state-of-the-art results for modeling and learning policies in Atari games as well as promising results in 2D LiDAR-based crowd navigation, can also be applied to the camera-based navigation problem. To this end, we create simulated environments where a robot must navigate past static and moving humans without colliding in order to reach its goal. We find that state-of-the-art methods are able to achieve success in solving the navigation problem, and can generate dream-like predictions of future image-sequences which show consistent geometry and moving persons. We are also able to show that policy performance in our high-fidelity sim2real simulation scenario transfers to the real world by testing the policy on a real robot. We make our simulator, models and experiments available at https://github.com/danieldugas/NavDreams.
The adoption of an appropriate approximate similarity search method is an essential prereq-uisite for developing a fast and efficient CBIR system, especially when dealing with large amount ofdata. In this study we implement a web image search engine on top of a Locality Sensitive Hashing(LSH) Index to allow fast similarity search on deep features. Specifically, we exploit transfer learningfor deep features extraction from images. Firstly, we adopt InceptionV3 pretrained on ImageNet asfeatures extractor, secondly, we try out several CNNs built on top of InceptionV3 as convolutionalbase fine-tuned on our dataset. In both of the previous cases we index the features extracted within ourLSH index implementation so as to compare the retrieval performances with and without fine-tuning.In our approach we try out two different LSH implementations: the first one working with real numberfeature vectors and the second one with the binary transposed version of those vectors. Interestingly,we obtain the best performances when using the binary LSH, reaching almost the same result, in termsof mean average precision, obtained by performing sequential scan of the features, thus avoiding thebias introduced by the LSH index. Lastly, we carry out a performance analysis class by class in terms ofrecall againstmAPhighlighting, as expected, a strong positive correlation between the two.
This work presents an analysis of state-of-the-art learning-based image compression techniques. We compare 8 models available in the Tensorflow Compression package in terms of visual quality metrics and processing time, using the KODAK data set. The results are compared with the Better Portable Graphics (BPG) and the JPEG2000 codecs. Results show that JPEG2000 has the lowest execution times compared with the fastest learning-based model, with a speedup of 1.46x in compression and 30x in decompression. However, the learning-based models achieved improvements over JPEG2000 in terms of quality, specially for lower bitrates. Our findings also show that BPG is more efficient in terms of PSNR, but the learning models are better for other quality metrics, and sometimes even faster. The results indicate that learning-based techniques are promising solutions towards a future mainstream compression method.
Existing deep learning-enabled semantic communication systems often rely on shared background knowledge between the transmitter and receiver that includes empirical data and their associated semantic information. In practice, the semantic information is defined by the pragmatic task of the receiver and cannot be known to the transmitter. The actual observable data at the transmitter can also have non-identical distribution with the empirical data in the shared background knowledge library. To address these practical issues, this paper proposes a new neural network-based semantic communication system for image transmission, where the task is unaware at the transmitter and the data environment is dynamic. The system consists of two main parts, namely the semantic extraction (SE) network and the data adaptation (DA) network. The SE network learns how to extract the semantic information using a receiver-leading training process. By using domain adaptation technique from transfer learning, the DA network learns how to convert the data observed into a similar form of the empirical data that the SE network can process without re-training. Numerical experiments show that the proposed method can be adaptive to observable datasets while keeping high performance in terms of both data recovery and task execution. The codes are available on https://github.com/SJTU-mxtao/Semantic-Communication-Systems.
Federated learning reduces the risk of information leakage, but remains vulnerable to attacks. We investigate how several neural network design decisions can defend against gradients inversion attacks. We show that overlapping gradients provides numerical resistance to gradient inversion on the highly vulnerable dense layer. Specifically, we propose to leverage batching to maximise mixing of gradients by choosing an appropriate loss function and drawing identical labels. We show that otherwise it is possible to directly recover all vectors in a mini-batch without any numerical optimisation due to the de-mixing nature of the cross entropy loss. To accurately assess data recovery, we introduce an absolute variation distance (AVD) metric for information leakage in images, derived from total variation. In contrast to standard metrics, e.g. Mean Squared Error or Structural Similarity Index, AVD offers a continuous metric for extracting information in noisy images. Finally, our empirical results on information recovery from various inversion attacks and training performance supports our defense strategies. These strategies are also shown to be useful for deep convolutional neural networks such as LeNET for image recognition. We hope that this study will help guide the development of further strategies that achieve a trustful federation policy.
Source device identification is an important topic in image forensics since it allows to trace back the origin of an image. Its forensics counter-part is source device anonymization, that is, to mask any trace on the image that can be useful for identifying the source device. A typical trace exploited for source device identification is the Photo Response Non-Uniformity (PRNU), a noise pattern left by the device on the acquired images. In this paper, we devise a methodology for suppressing such a trace from natural images without significant impact on image quality. Specifically, we turn PRNU anonymization into an optimization problem in a Deep Image Prior (DIP) framework. In a nutshell, a Convolutional Neural Network (CNN) acts as generator and returns an image that is anonymized with respect to the source PRNU, still maintaining high visual quality. With respect to widely-adopted deep learning paradigms, our proposed CNN is not trained on a set of input-target pairs of images. Instead, it is optimized to reconstruct the PRNU-free image from the original image under analysis itself. This makes the approach particularly suitable in scenarios where large heterogeneous databases are analyzed and prevents any problem due to lack of generalization. Through numerical examples on publicly available datasets, we prove our methodology to be effective compared to state-of-the-art techniques.
Proposed in 2014, Generative Adversarial Networks (GAN) initiated a fresh interest in generative modelling. They immediately achieved state-of-the-art in image synthesis, image-to-image translation, text-to-image generation, image inpainting and have been used in sciences ranging from medicine to high-energy particle physics. Despite their popularity and ability to learn arbitrary distributions, GAN have not been widely applied in recommender systems (RS). Moreover, only few of the techniques that have introduced GAN in RS have employed them directly as a collaborative filtering (CF) model. In this work we propose a new GAN-based approach that learns user and item latent factors in a matrix factorization setting for the generic top-N recommendation problem. Following the vector-wise GAN training approach for RS introduced by CFGAN, we identify 2 unique issues when utilizing GAN for CF. We propose solutions for both of them by using an autoencoder as discriminator and incorporating an additional loss function for the generator. We evaluate our model, GANMF, through well-known datasets in the RS community and show improvements over traditional CF approaches and GAN-based models. Through an ablation study on the components of GANMF we aim to understand the effects of our architectural choices. Finally, we provide a qualitative evaluation of the matrix factorization performance of GANMF.
We study the query-based attack against image retrieval to evaluate its robustness against adversarial examples under the black-box setting, where the adversary only has query access to the top-k ranked unlabeled images from the database. Compared with query attacks in image classification, which produce adversaries according to the returned labels or confidence score, the challenge becomes even more prominent due to the difficulty in quantifying the attack effectiveness on the partial retrieved list. In this paper, we make the first attempt in Query-based Attack against Image Retrieval (QAIR), to completely subvert the top-k retrieval results. Specifically, a new relevance-based loss is designed to quantify the attack effects by measuring the set similarity on the top-k retrieval results before and after attacks and guide the gradient optimization. To further boost the attack efficiency, a recursive model stealing method is proposed to acquire transferable priors on the target model and generate the prior-guided gradients. Comprehensive experiments show that the proposed attack achieves a high attack success rate with few queries against the image retrieval systems under the black-box setting. The attack evaluations on the real-world visual search engine show that it successfully deceives a commercial system such as Bing Visual Search with 98% attack success rate by only 33 queries on average.
Point clouds have the characteristics of disorder, unstructured and sparseness.Aiming at the problem of the non-structural nature of point clouds, thanks to the excellent performance of convolutional neural networks in image processing, one of the solutions is to extract features from point clouds based on two-dimensional convolutional neural networks. The three-dimensional information carried in the point cloud can be converted to two-dimensional, and then processed by a two-dimensional convolutional neural network, and finally back-projected to three-dimensional.In the process of projecting 3D information to 2D and back-projection, certain information loss will inevitably be caused to the point cloud and category inconsistency will be introduced in the back-projection stage;Another solution is the voxel-based point cloud segmentation method, which divides the point cloud into small grids one by one.However, the point cloud is sparse, and the direct use of 3D convolutional neural network inevitably wastes computing resources. In this paper, we propose a feature extraction module based on multi-scale ultra-sparse convolution and a feature selection module based on channel attention, and build a point cloud segmentation network framework based on this.By introducing multi-scale sparse convolution, network could capture richer feature information based on convolution kernels of different sizes, improving the segmentation result of point cloud segmentation.
Temporal alignment of fine-grained human actions in videos is important for numerous applications in computer vision, robotics, and mixed reality. State-of-the-art methods directly learn image-based embedding space by leveraging powerful deep convolutional neural networks. While being straightforward, their results are far from satisfactory, the aligned videos exhibit severe temporal discontinuity without additional post-processing steps. The recent advancements in human body and hand pose estimation in the wild promise new ways of addressing the task of human action alignment in videos. In this work, based on off-the-shelf human pose estimators, we propose a novel context-aware self-supervised learning architecture to align sequences of actions. We name it CASA. Specifically, CASA employs self-attention and cross-attention mechanisms to incorporate the spatial and temporal context of human actions, which can solve the temporal discontinuity problem. Moreover, we introduce a self-supervised learning scheme that is empowered by novel 4D augmentation techniques for 3D skeleton representations. We systematically evaluate the key components of our method. Our experiments on three public datasets demonstrate CASA significantly improves phase progress and Kendall's Tau scores over the previous state-of-the-art methods.