We propose a transformer-based neural network architecture for multi-object 3D reconstruction from RGB videos. It relies on two alternative ways to represent its knowledge: as a global 3D grid of features and an array of view-specific 2D grids. We progressively exchange information between the two with a dedicated bidirectional attention mechanism. We exploit knowledge about the image formation process to significantly sparsify the attention weight matrix, making our architecture feasible on current hardware, both in terms of memory and computation. We attach a DETR-style head on top of the 3D feature grid in order to detect the objects in the scene and to predict their 3D pose and 3D shape. Compared to previous methods, our architecture is single stage, end-to-end trainable, and it can reason holistically about a scene from multiple video frames without needing a brittle tracking step. We evaluate our method on the challenging Scan2CAD dataset, where we outperform (1) recent state-of-the-art methods for 3D object pose estimation from RGB videos; and (2) a strong alternative method combining Multi-view Stereo with RGB-D CAD alignment. We plan to release our source code.
Quantum Machine Learning models are composed by Variational Quantum Circuits (VQCs) in a very natural way. There are already some empirical results proving that such models provide an advantage in supervised/unsupervised learning tasks. However, when applied to Reinforcement Learning (RL), less is known. In this work, we consider Policy Gradients using a hardware-efficient ansatz. We prove that the complexity of obtaining an {\epsilon}-approximation of the gradient using quantum hardware scales only logarithmically with the number of parameters, considering the number of quantum circuits executions. We test the performance of such models in benchmarking environments and verify empirically that such quantum models outperform typical classical neural networks used in those environments, using a fraction of the number of parameters. Moreover, we propose the utilization of the Fisher Information spectrum to show that the quantum model is less prone to barren plateaus than its classical counterpart. As a different use case, we consider the application of such variational quantum models to the problem of quantum control and show its feasibility in the quantum-quantum domain.
Semi- and weakly-supervised learning have recently attracted considerable attention in the object detection literature since they can alleviate the cost of annotation needed to successfully train deep learning models. State-of-art approaches for semi-supervised learning rely on student-teacher models trained using a multi-stage process, and considerable data augmentation. Custom networks have been developed for the weakly-supervised setting, making it difficult to adapt to different detectors. In this paper, a weakly semi-supervised training method is introduced that reduces these training challenges, yet achieves state-of-the-art performance by leveraging only a small fraction of fully-labeled images with information in weakly-labeled images. In particular, our generic sampling-based learning strategy produces pseudo-ground-truth (GT) bounding box annotations in an online fashion, eliminating the need for multi-stage training, and student-teacher network configurations. These pseudo GT boxes are sampled from weakly-labeled images based on the categorical score of object proposals accumulated via a score propagation process. Empirical results on the Pascal VOC dataset, indicate that the proposed approach improves performance by 5.0% when using VOC 2007 as fully-labeled, and VOC 2012 as weak-labeled data. Also, with 5-10% fully annotated images, we observed an improvement of more than 10% in mAP, showing that a modest investment in image-level annotation, can substantially improve detection performance.
In our previous work, we proposed a discriminative autoencoder (DcAE) for speech recognition. DcAE combines two training schemes into one. First, since DcAE aims to learn encoder-decoder mappings, the squared error between the reconstructed speech and the input speech is minimized. Second, in the code layer, frame-based phonetic embeddings are obtained by minimizing the categorical cross-entropy between ground truth labels and predicted triphone-state scores. DcAE is developed based on the Kaldi toolkit by treating various TDNN models as encoders. In this paper, we further propose three new versions of DcAE. First, a new objective function that considers both categorical cross-entropy and mutual information between ground truth and predicted triphone-state sequences is used. The resulting DcAE is called a chain-based DcAE (c-DcAE). For application to robust speech recognition, we further extend c-DcAE to hierarchical and parallel structures, resulting in hc-DcAE and pc-DcAE. In these two models, both the error between the reconstructed noisy speech and the input noisy speech and the error between the enhanced speech and the reference clean speech are taken into the objective function. Experimental results on the WSJ and Aurora-4 corpora show that our DcAE models outperform baseline systems.
Movie genre classification has been widely studied in recent years due to its various applications in video editing, summarization, and recommendation. Prior work has typically addressed this task by predicting genres based solely on the visual content. As a result, predictions from these methods often perform poorly for genres such as documentary or musical, since non-visual modalities like audio or language play an important role in correctly classifying these genres. In addition, the analysis of long videos at frame level is always associated with high computational cost and makes the prediction less efficient. To address these two issues, we propose a Multi-Modal approach leveraging shot information, MMShot, to classify video genres in an efficient and effective way. We evaluate our method on MovieNet and Condensed Movies for genre classification, achieving 17% ~ 21% improvement on mean Average Precision (mAP) over the state-of-the-art. Extensive experiments are conducted to demonstrate the ability of MMShot for long video analysis and uncover the correlations between genres and multiple movie elements. We also demonstrate our approach's ability to generalize by evaluating the scene boundary detection task, achieving 1.1% improvement on Average Precision (AP) over the state-of-the-art.
Recently, supervised network embedding (NE) has emerged as a predominant technique for representing complex systems that take the form of networks, and various downstream node- and network-level tasks have benefited from its remarkable developments. However, unsupervised NE still remains challenging due to the uncertainty in defining a learning objective. In addition, it is still an unexplored research question whether existing NE methods adapt well to heterophilous networks. This paper introduces the first empirical study on the influence of homophily ratio on the performance of existing unsupervised NE methods and reveals their limitations. Inspired by our empirical findings, we design unsupervised NE task as an r-ego network discrimination problem and further develop a SELf-supErvised Network Embedding (Selene) framework for learning useful node representations for both homophilous and heterophilous networks. Specifically, we propose a dual-channel feature embedding mechanism to fuse node attributes and network structure information and leverage a sampling and anonymisation strategy to break the implicit homophily assumption of existing embedding mechanisms. Lastly, we introduce a negative-sample-free SSL objective function to optimise the framework. We conduct extensive experiments and a series of ablation studies on 12 real-world datasets and 20 synthetic networks. Results demonstrate Selene's superior performance and confirm the effectiveness of each component.
Contrastive learning relies on an assumption that positive pairs contain related views, e.g., patches of an image or co-occurring multimodal signals of a video, that share certain underlying information about an instance. But what if this assumption is violated? The literature suggests that contrastive learning produces suboptimal representations in the presence of noisy views, e.g., false positive pairs with no apparent shared information. In this work, we propose a new contrastive loss function that is robust against noisy views. We provide rigorous theoretical justifications by showing connections to robust symmetric losses for noisy binary classification and by establishing a new contrastive bound for mutual information maximization based on the Wasserstein distance measure. The proposed loss is completely modality-agnostic and a simple drop-in replacement for the InfoNCE loss, which makes it easy to apply to existing contrastive frameworks. We show that our approach provides consistent improvements over the state-of-the-art on image, video, and graph contrastive learning benchmarks that exhibit a variety of real-world noise patterns.
Some neurons in deep networks specialize in recognizing highly specific perceptual, structural, or semantic features of inputs. In computer vision, techniques exist for identifying neurons that respond to individual concept categories like colors, textures, and object classes. But these techniques are limited in scope, labeling only a small subset of neurons and behaviors in any network. Is a richer characterization of neuron-level computation possible? We introduce a procedure (called MILAN, for mutual-information-guided linguistic annotation of neurons) that automatically labels neurons with open-ended, compositional, natural language descriptions. Given a neuron, MILAN generates a description by searching for a natural language string that maximizes pointwise mutual information with the image regions in which the neuron is active. MILAN produces fine-grained descriptions that capture categorical, relational, and logical structure in learned features. These descriptions obtain high agreement with human-generated feature descriptions across a diverse set of model architectures and tasks, and can aid in understanding and controlling learned models. We highlight three applications of natural language neuron descriptions. First, we use MILAN for analysis, characterizing the distribution and importance of neurons selective for attribute, category, and relational information in vision models. Second, we use MILAN for auditing, surfacing neurons sensitive to protected categories like race and gender in models trained on datasets intended to obscure these features. Finally, we use MILAN for editing, improving robustness in an image classifier by deleting neurons sensitive to text features spuriously correlated with class labels.
Computationally expensive neural networks are ubiquitous in computer vision and solutions for efficient inference have drawn a growing attention in the machine learning community. Examples of such solutions comprise quantization, i.e. converting the processing values (weights and inputs) from floating point into integers e.g. int8 or int4. Concurrently, the rise of privacy concerns motivated the study of less invasive acceleration methods, such as data-free quantization of pre-trained models weights and activations. Previous approaches either exploit statistical information to deduce scalar ranges and scaling factors for the activations in a static manner, or dynamically adapt this range on-the-fly for each input of each layers (also referred to as activations): the latter generally being more accurate at the expanse of significantly slower inference. In this work, we argue that static input quantization can reach the accuracy levels of dynamic methods by means of a per-channel input quantization scheme that allows one to more finely preserve cross-channel dynamics. We show through a thorough empirical evaluation on multiple computer vision problems (e.g. ImageNet classification, Pascal VOC object detection as well as CityScapes semantic segmentation) that the proposed method, dubbed SPIQ, achieves accuracies rivalling dynamic approaches with static-level inference speed, significantly outperforming state-of-the-art quantization methods on every benchmark.
This paper presents an analysis of speech synthesis quality achieved by simultaneously performing voice conversion and language code-switching using multilingual VQ-VAE speech synthesis in German, French, English and Italian. In this paper, we utilize VQ code indices representing phone information from VQ-VAE to perform code-switching and a VQ speaker code to perform voice conversion in a single system with a neural vocoder. Our analysis examines several aspects of code-switching including the number of language switches and the number of words involved in each switch. We found that speech synthesis quality degrades after increasing the number of language switches within an utterance and decreasing the number of words. We also found some evidence of accent transfer when performing voice conversion across languages as observed when a speaker's original language differs from the language of a synthetic target utterance. We present results from our listening tests and discuss the inherent difficulties of assessing accent transfer in speech synthesis. Our work highlights some of the limitations and strengths of using a semi-supervised end-to-end system like VQ-VAE for handling multilingual synthesis. Our work provides insight into why multilingual speech synthesis is challenging and we suggest some directions for expanding work in this area.