Abstract

In this paper, we investigate a constrained formulation of neural networks where the output is a convex function of the input. We show that the convexity constraints can be enforced on both fully connected and convolutional layers, making them applicable to most architectures. The convexity constraints include restricting the weights (for all but the first layer) to be non-negative and using a non-decreasing convex activation function. Albeit simple, these constraints have profound implications on the generalization abilities of the network. We draw three valuable insights: (a) Input Output Convex Networks (IOC-NN) self regularize and almost uproot the problem of overfitting; (b) Although heavily constrained, they come close to the performance of the base architectures; and (c) The ensemble of convex networks can match or outperform the non convex counterparts. We demonstrate the efficacy of the proposed idea using thorough experiments and ablation studies on MNIST, CIFAR10, and CIFAR100 datasets with three different neural network architectures. The code for this project is publicly available at: https://github.com/sarathsp1729/Convex-Networks.

Abstract

We propose the ViNet architecture for audiovisual saliency prediction. ViNet is a fully convolutional encoder-decoder architecture. The encoder uses visual features from a network trained for action recognition, and the decoder infers a saliency map via trilinear interpolation and 3D convolutions, combining features from multiple hierarchies. The overall architecture of ViNet is conceptually simple; it is causal and runs in real-time (60 fps). ViNet does not use audio as input and still outperforms the state-of-the-art audio-visual saliency prediction models on nine different datasets (three visual-only and six audio-visual datasets). ViNet also surpasses human performance on the CC, SIM and AUC metrics for the AVE dataset, and to our knowledge, it is the first model to do so. We also explore a variation of ViNet architecture by augmenting audio features into the decoder. To our surprise, upon sufficient training, the network becomes agnostic to the input audio and provides the same output irrespective of the input. Interestingly, we also observe similar behaviour in the previous state-of-the-art models [1] for audio-visual saliency prediction. Our findings contrast with previous works on deep learning-based audio-visual saliency prediction, suggesting a clear avenue for future explorations incorporating audio in a more effective manner. The code and pre-trained models are

Abstract

We propose thetextbf {AViNet} architecture for audiovisual saliency prediction. AViNet is a fully convolutional encoder-decoder architecture. The encoder combines visual features learned for action recognition, with audio embeddings learned via an aural network designed to classify objects and scenes. The decoder infers a saliency map via trilinear interpolation and 3D convolutions, combining hierarchical features. The overall architecture is conceptually simple, causal, and runs in real-time (60 fps). AViNet outperforms the state-of-the-art on ten (seven audiovisual and three visual-only) datasets while surpassing human performance on the CC, SIM, and AUC metrics for the AVE dataset. Visual features maximally account for saliency on existing datasets with audio-only contributing to minor gains, except in specific contexts like social events. Our work, therefore, motivates the need to curate saliency datasets reflective of real-life, where both the visual and aural modalities complimentarily drive saliency. Our code and pre-trained models are available at this https URL

Abstract

In one aspect, a computerized method for automatically generating digital video content from scripts includes a media engine. The media engine receives a script and a plurality of flags; sending the script to a natural language processing (NLP) engine. The NLP engine parses the script. The script is broken into a set of keywords and phrases in a JSON format. The NLP engine, based on the keywords and phrases and the plurality of flags, obtains a relevant background scene for the video and a relevant set of assets for the video, and a set of associated attributes of each of the set of assets. In an asset is a character or object for the video. The NLP engine provides the parsed script in the JSON format and the relevant background scene of the video and the relevant set of assets of the video, and the set of associated attributes of each of the set of assets to a layout engine. The layout engine, based on the parsed script …

Abstract

There has been increasing interest in building deep hierarchy-aware classifiers that aim to quantify and reduce the severity of mistakes, and not just reduce the number of errors. The idea is to exploit the label hierarchy (e.g., the WordNet ontology) and consider graph distances as a proxy for mistake severity. Surprisingly, on examining mistake-severity distributions of the top-1 prediction, we find that current state-of-the-art hierarchy-aware deep classifiers do not always show practical improvement over the standard cross-entropy baseline in making better mistakes. The reason for the reduction in average mistake-severity can be attributed to the increase in low-severity mistakes, which may also explain the noticeable drop in their accuracy. To this end, we use the classical Conditional Risk Minimization (CRM) framework for hierarchy-aware classification. Given a cost matrix and a reliable estimate of likelihoods (obtained from a trained network), CRM simply amends mistakes at inference time; it needs no extra hyperparameters and requires adding just a few lines of code to the standard cross-entropy baseline. It significantly outperforms the state-of-the-art and consistently obtains large reductions in the average hierarchical distance of top- predictions across datasets, with very little loss in accuracy. CRM, because of its simplicity, can be used with any off-the-shelf trained model that provides reliable likelihood estimates.

Abstract

— Humans have a natural ability to effortlessly comprehend linguistic commands such as “park next to the yellow sedan” and instinctively know which region of the road the vehicle should navigate. Extending this ability to autonomous vehicles is the next step towards creating fully autonomous agents that respond and act according to human commands. To this end, we propose the novel task of Referring Navigable Regions (RNR), i.e., grounding regions of interest for navigation based on the linguistic command. RNR is different from Referring Image Segmentation (RIS), which focuses on grounding an object referred to by the natural language expression instead of grounding a navigable region. For example, for a command “park next to the yellow sedan,” RIS will aim to segment the referred sedan, and RNR aims to segment the suggested parking region on the road. We introduce a new dataset, Talk2Car-RegSeg, which extends the existing Talk2car [1] dataset with segmentation masks for the regions described by the linguistic commands. A separate test split with concise manoeuvre-oriented commands is provided to assess the practicality of our dataset. We benchmark the proposed dataset using a novel transformer-based architecture. We present extensive ablations and show superior performance over baselines on multiple evaluation metrics. A downstream path planner generating trajectories based on RNR outputs confirms the efficacy of the proposed framework.

Abstract

No Results Found

Abstract

Machine-generated speech is characterized by its limited or unnatural emotional variation. Current text to speech systems generates speech with either a flat emotion, emotion selected from a predefined set, average variation learned from prosody sequences in training data or transferred from a source style. We propose a text to speech(TTS) system, where a user can choose the emotion of generated speech from a continuous and meaningful emotion space (Arousal-Valence space). The proposed TTS system can generate speech from the text in any speaker’s style, with fine control of emotion. We show that the system works on emotion unseen during training and can scale to previously unseen speakers given his/her speech sample. Our work expands the horizon of the state-of-the-art FastSpeech2 backbone to a multi-speaker setting and gives it much-coveted continuous (and interpretable) affective control, without any observable degradation in the quality of the synthesized speech. Audio samples are available at https://researchweb.iiit.ac. in/˜sarath.s/emotts/

Abstract

main generalization (DG) of machine learning algorithms is defined as their ability to learn a domain agnostic hypothesis from multiple training distributions, which generalizes onto data from an unseen domain. DG is vital in scenarios where the target domain with distinct characteristics has sparse data for training. Aligning with recent work [9], we find that a straightforward Empirical Risk Minimization (ERM) baseline consistently outperforms existing DG methods. We present ablation studies indicating that the choice of backbone, data augmentation, and optimization algorithms overshadows the many tricks and trades explored in the prior art. Our work leads to a new state of the art on the four popular DG datasets, surpassing previous methods by large margins. Furthermore, as a key contribution, we propose a classwise-DG formulation, where for each class, we randomly select one of the domains and keep it aside for testing. We argue that this benchmarking is closer to human learning and relevant in real-world scenarios. We comprehensively benchmark classwise-DG on the DomainBed [9] and propose a method combining ERM and reverse gradients to achieve the state-of-the-art results. To our surprise, despite being exposed to all domains during training, the classwise DG is more challenging than traditional DG evaluation and motivates more fundamental rethinking on the problem of DG.

Abstract

There has been increasing interest in building deep hierarchy-aware classifiers that aim to quantify and reduce the severity of mistakes, and not just reduce the number of errors. The idea is to exploit the label hierarchy (e.g., the WordNet ontology) and consider graph distances as a proxy for mistake severity. Surprisingly, on examining mistake-severity distributions of the top-1 prediction, we find that current state-of-the-art hierarchy-aware deep classifiers do not always show practical improvement over the standard cross-entropy baseline in making better mistakes. The reason for the reduction in average mistake-severity can be attributed to the increase in low-severity mistakes, which may also explain the noticeable drop in their accuracy. To this end, we use the classical Conditional Risk Minimization (CRM) framework for hierarchy-aware classification. Given a cost matrix and a reliable estimate of likelihoods (obtained from a trained network), CRM simply amends mistakes at inference time; it needs no extra hyperparameters and requires adding just a few lines of code to the standard cross-entropy baseline. It significantly outperforms the state-of-the-art and consistently obtains large reductions in the average hierarchical distance of top- predictions across datasets, with very little loss in accuracy. CRM, because of its simplicity, can be used with any off-the-shelf trained model that provides reliable likelihood estimates.