We address the problem of text-guided video temporal grounding, which aims to identify the time interval of certain event based on a natural language description. Different from most existing methods that only consider RGB images as visual features, we propose a multi-modal framework to extract complementary information from videos. Specifically, we adopt RGB images for appearance, optical flow for motion, and depth maps for image structure. While RGB images provide abundant visual cues of certain event, the performance may be affected by background clutters. Therefore, we use optical flow to focus on large motion and depth maps to infer the scene configuration when the action is related to objects recognizable with their shapes. To integrate the three modalities more effectively and enable inter-modal learning, we design a dynamic fusion scheme with transformers to model the interactions between modalities. Furthermore, we apply intra-modal self-supervised learning to enhance feature representations across videos for each modality, which also facilitates multi-modal learning. We conduct extensive experiments on the Charades-STA and ActivityNet Captions datasets, and show that the proposed method performs favorably against state-of-the-art approaches.
We introduce the new Birds-to-Words dataset of 41k sentences describing fine-grained differences between photographs of birds. The language collected is highly detailed, while remaining understandable to the everyday observer (e.g., "heart-shaped face," "squat body"). Paragraph-length descriptions naturally adapt to varying levels of taxonomic and visual distance---drawn from a novel stratified sampling approach---with the appropriate level of detail. We propose a new model called Neural Naturalist that uses a joint image encoding and comparative module to generate comparative language, and evaluate the results with humans who must use the descriptions to distinguish real images. Our results indicate promising potential for neural models to explain differences in visual embedding space using natural language, as well as a concrete path for machine learning to aid citizen scientists in their effort to preserve biodiversity.
Advances in object-centric generative models (OCGMs) have culminated in the development of a broad range of methods for unsupervised object segmentation and interpretable object-centric scene generation. These methods, however, are limited to simulated and real-world datasets with limited visual complexity. Moreover, object representations are often inferred using RNNs which do not scale well to large images or iterative refinement which avoids imposing an unnatural ordering on objects in an image but requires the a priori initialisation of a fixed number of object representations. In contrast to established paradigms, this work proposes an embedding-based approach in which embeddings of pixels are clustered in a differentiable fashion using a stochastic, non-parametric stick-breaking process. Similar to iterative refinement, this clustering procedure also leads to randomly ordered object representations, but without the need of initialising a fixed number of clusters a priori. This is used to develop a new model, GENESIS-V2, which can infer a variable number of object representations without using RNNs or iterative refinement. We show that GENESIS-V2 outperforms previous methods for unsupervised image segmentation and object-centric scene generation on established synthetic datasets as well as more complex real-world datasets.
Existing psychophysical studies have revealed that the cross-modal visual-tactile perception is common for humans performing daily activities. However, it is still challenging to build the algorithmic mapping from one modality space to another, namely the cross-modal visual-tactile data translation/generation, which could be potentially important for robotic operation. In this paper, we propose a deep-learning-based approach for cross-modal visual-tactile data generation by leveraging the framework of the generative adversarial networks (GANs). Our approach takes the visual image of a material surface as the visual data, and the accelerometer signal induced by the pen-sliding movement on the surface as the tactile data. We adopt the conditional-GAN (cGAN) structure together with the residue-fusion (RF) module, and train the model with the additional feature-matching (FM) and perceptual losses to achieve the cross-modal data generation. The experimental results show that the inclusion of the RF module, and the FM and the perceptual losses significantly improves cross-modal data generation performance in terms of the classification accuracy upon the generated data and the visual similarity between the ground-truth and the generated data.
This concept paper aims to provide a brief outline of quantum computers, explore existing methods of quantum image classification techniques, so focusing on remote sensing applications, and discuss the bottlenecks of performing these algorithms on currently available open source platforms. Initial results demonstrate feasibility. Next steps include expanding the size of the quantum hidden layer and increasing the variety of output image options.
Generating natural and accurate descriptions in image cap-tioning has always been a challenge. In this paper, we pro-pose a novel recall mechanism to imitate the way human con-duct captioning. There are three parts in our recall mecha-nism : recall unit, semantic guide (SG) and recalled-wordslot (RWS). Recall unit is a text-retrieval module designedto retrieve recalled words for images. SG and RWS are de-signed for the best use of recalled words. SG branch cangenerate a recalled context, which can guide the process ofgenerating caption. RWS branch is responsible for copyingrecalled words to the caption. Inspired by pointing mecha-nism in text summarization, we adopt a soft switch to balancethe generated-word probabilities between SG and RWS. Inthe CIDEr optimization step, we also introduce an individualrecalled-word reward (WR) to boost training. Our proposedmethods (SG+RWS+WR) achieve BLEU-4 / CIDEr / SPICEscores of 36.6 / 116.9 / 21.3 with cross-entropy loss and 38.7 /129.1 / 22.4 with CIDEr optimization on MSCOCO Karpathytest split, which surpass the results of other state-of-the-artmethods.
It is a challenging task to restore images from their variants with combined distortions. In the existing works, a promising strategy is to apply parallel "operations" to handle different types of distortion. However, in the feature fusion phase, a small number of operations would dominate the restoration result due to the features' heterogeneity by different operations. To this end, we introduce the tensor 1x1 convolutional layer by imposing high-order tensor (outer) product, by which we not only harmonize the heterogeneous features but also take additional non-linearity into account. To avoid the unacceptable kernel size resulted from the tensor product, we construct the kernels with tensor network decomposition, which is able to convert the exponential growth of the dimension to linear growth. Armed with the new layer, we propose High-order OWAN for multi-distorted image restoration. In the numerical experiments, the proposed net outperforms the previous state-of-the-art and shows promising performance even in more difficult tasks.
The power-balanced hybrid optical imaging system is a special design of a computational camera, introduced in this paper, with image formation by a refractive lens and Multilevel Phase Mask (MPM) as a diffractive optical element (DoE). This system provides a long focal depth and low chromatic aberrations thanks to MPM, and a high energy light concentration due to the refractive lens. This paper additionally introduces the concept of a optimal power balance between lens and MPM for achromatic extended-depth-of-field (EDoF) imaging. To optimize this power-balance as well as to optimize MPM using Neural Network techniques, we build a fully-differentiable image formation model for joint optimization of optical and imaging parameters for the designed computational camera. Additionally, we determine a Wiener-like inverse imaging optimal optical transfer function (OTF) to reconstruct a sharp image from the defocused observation. We numerically and experimentally compare the designed system with its counterparts, lensless and just-lens optical systems, for the visible wavelength interval (400-700) nm and the EDoF range (0.5-1000) m. The attained results demonstrate that the proposed system equipped with the optimal OTF overcomes its lensless and just-lens counterparts (even when they are used with optimized OTFs) in terms of reconstruction quality for off-focus distances.
Transferring image-based object detectors to the domain of video remains challenging under resource constraints. Previous efforts utilised optical flow to allow unchanged features to be propagated, however, the overhead is considerable when working with very slowly changing scenes from applications such as surveillance. In this paper, we propose temporal early exits to reduce the computational complexity of per-frame video object detection. Multiple temporal early exit modules with low computational overhead are inserted at early layers of the backbone network to identify the semantic differences between consecutive frames. Full computation is only required if the frame is identified as having a semantic change to previous frames; otherwise, detection results from previous frames are reused. Experiments on CDnet show that our method significantly reduces the computational complexity and execution of per-frame video object detection up to $34 \times$ compared to existing methods with an acceptable reduction of 2.2\% in mAP.
Convolutional neural networks (CNNs) learn to extract representations of complex features, such as object shapes and textures to solve image recognition tasks. Recent work indicates that CNNs trained on ImageNet are biased towards features that encode textures and that these alone are sufficient to generalize to unseen test data from the same distribution as the training data but often fail to generalize to out-of-distribution data. It has been shown that augmenting the training data with different image styles decreases this texture bias in favor of increased shape bias while at the same time improving robustness to common corruptions, such as noise and blur. Commonly, this is interpreted as shape bias increasing corruption robustness. However, this relationship is only hypothesized. We perform a systematic study of different ways of composing inputs based on natural images, explicit edge information, and stylization. While stylization is essential for achieving high corruption robustness, we do not find a clear correlation between shape bias and robustness. We conclude that the data augmentation caused by style-variation accounts for the improved corruption robustness and increased shape bias is only a byproduct.