arXiv subjects

Dongxu Li

Publications and source records attributed to Dongxu Li.

47 records · Page 3Linked to original sources

arXiv2021

Align and Prompt: Video-and-Language Pre-training with Entity Prompts

Video-and-language pre-training has shown promising improvements on various downstream tasks. Most previous methods capture cross-modal interactions with a transformer-based multimodal encoder, not fully addressing the misalignment between unimodal video and text features. Besides, learning fine-grained visual-language alignment usually requires off-the-shelf object detectors to provide object information, which is bottlenecked by the detector's limited vocabulary and expensive computation cost. We propose Align and Prompt: an efficient and effective video-and-language pre-training framework with better cross-modal alignment. First, we introduce a video-text contrastive (VTC) loss to align unimodal video-text features at the instance level, which eases the modeling of cross-modal interactions. Then, we propose a new visually-grounded pre-training task, prompting entity modeling (PEM), which aims to learn fine-grained region-entity alignment. To achieve this, we first introduce an entity prompter module, which is trained with VTC to produce the similarity between a video crop and text prompts instantiated with entity names. The PEM task then asks the model to predict the entity pseudo-labels (i.e~normalized similarity scores) for randomly-selected video crops. The resulting pre-trained model achieves state-of-the-art performance on both text-video retrieval and videoQA, outperforming prior work by a substantial margin. Our code and pre-trained models are available at https://github.com/salesforce/ALPRO.

cs.CV↗

arXiv2021

Transcribing Natural Languages for The Deaf via Neural Editing Programs

This work studies the task of glossification, of which the aim is to em transcribe natural spoken language sentences for the Deaf (hard-of-hearing) community to ordered sign language glosses. Previous sequence-to-sequence language models trained with paired sentence-gloss data often fail to capture the rich connections between the two distinct languages, leading to unsatisfactory transcriptions. We observe that despite different grammars, glosses effectively simplify sentences for the ease of deaf communication, while sharing a large portion of vocabulary with sentences. This has motivated us to implement glossification by executing a collection of editing actions, e.g. word addition, deletion, and copying, called editing programs, on their natural spoken language counterparts. Specifically, we design a new neural agent that learns to synthesize and execute editing programs, conditioned on sentence contexts and partial editing results. The agent is trained to imitate minimal editing programs, while exploring more widely the program space via policy gradients to optimize sequence-wise transcription quality. Results show that our approach outperforms previous glossification models by a large margin.

cs.CL↗

arXiv2021

Dual Attention-in-Attention Model for Joint Rain Streak and Raindrop Removal

Rain streaks and rain drops are two natural phenomena, which degrade image capture in different ways. Currently, most existing deep deraining networks take them as two distinct problems and individually address one, and thus cannot deal adequately with both simultaneously. To address this, we propose a Dual Attention-in-Attention Model (DAiAM) which includes two DAMs for removing both rain streaks and raindrops. Inside the DAM, there are two attentive maps - each of which attends to the heavy and light rainy regions, respectively, to guide the deraining process differently for applicable regions. In addition, to further refine the result, a Differential-driven Dual Attention-in-Attention Model (D-DAiAM) is proposed with a "heavy-to-light" scheme to remove rain via addressing the unsatisfying deraining regions. Extensive experiments on one public raindrop dataset, one public rain streak and our synthesized joint rain streak and raindrop (JRSRD) dataset have demonstrated that the proposed method not only is capable of removing rain streaks and raindrops simultaneously, but also achieves the state-of-the-art performance on both tasks.

cs.CV↗

arXiv2021

ARVo: Learning All-Range Volumetric Correspondence for Video Deblurring

Video deblurring models exploit consecutive frames to remove blurs from camera shakes and object motions. In order to utilize neighboring sharp patches, typical methods rely mainly on homography or optical flows to spatially align neighboring blurry frames. However, such explicit approaches are less effective in the presence of fast motions with large pixel displacements. In this work, we propose a novel implicit method to learn spatial correspondence among blurry frames in the feature space. To construct distant pixel correspondences, our model builds a correlation volume pyramid among all the pixel-pairs between neighboring frames. To enhance the features of the reference frame, we design a correlative aggregation module that maximizes the pixel-pair correlations with its neighbors based on the volume pyramid. Finally, we feed the aggregated features into a reconstruction module to obtain the restored frame. We design a generative adversarial paradigm to optimize the model progressively. Our proposed method is evaluated on the widely-adopted DVD dataset, along with a newly collected High-Frame-Rate (1000 fps) Dataset for Video Deblurring (HFR-DVD). Quantitative and qualitative experiments show that our model performs favorably on both datasets against previous state-of-the-art methods, confirming the benefit of modeling all-range spatial correspondence for video deblurring.

cs.CV↗

arXiv2020

TSPNet: Hierarchical Feature Learning via Temporal Semantic Pyramid for Sign Language Translation

Sign language translation (SLT) aims to interpret sign video sequences into text-based natural language sentences. Sign videos consist of continuous sequences of sign gestures with no clear boundaries in between. Existing SLT models usually represent sign visual features in a frame-wise manner so as to avoid needing to explicitly segmenting the videos into isolated signs. However, these methods neglect the temporal information of signs and lead to substantial ambiguity in translation. In this paper, we explore the temporal semantic structures of signvideos to learn more discriminative features. To this end, we first present a novel sign video segment representation which takes into account multiple temporal granularities, thus alleviating the need for accurate video segmentation. Taking advantage of the proposed segment representation, we develop a novel hierarchical sign video feature learning method via a temporal semantic pyramid network, called TSPNet. Specifically, TSPNet introduces an inter-scale attention to evaluate and enhance local semantic consistency of sign segments and an intra-scale attention to resolve semantic ambiguity by using non-local video context. Experiments show that our TSPNet outperforms the state-of-the-art with significant improvements on the BLEU score (from 9.58 to 13.41) and ROUGE score (from 31.80 to 34.96)on the largest commonly-used SLT dataset. Our implementation is available at https://github.com/verashira/TSPNet.

cs.CV↗

arXiv2020

Transferring Cross-domain Knowledge for Video Sign Language Recognition

Word-level sign language recognition (WSLR) is a fundamental task in sign language interpretation. It requires models to recognize isolated sign words from videos. However, annotating WSLR data needs expert knowledge, thus limiting WSLR dataset acquisition. On the contrary, there are abundant subtitled sign news videos on the internet. Since these videos have no word-level annotation and exhibit a large domain gap from isolated signs, they cannot be directly used for training WSLR models. We observe that despite the existence of a large domain gap, isolated and news signs share the same visual concepts, such as hand gestures and body movements. Motivated by this observation, we propose a novel method that learns domain-invariant visual concepts and fertilizes WSLR models by transferring knowledge of subtitled news sign to them. To this end, we extract news signs using a base WSLR model, and then design a classifier jointly trained on news and isolated signs to coarsely align these two domain features. In order to learn domain-invariant features within each class and suppress domain-specific features, our method further resorts to an external memory to store the class centroids of the aligned news signs. We then design a temporal attention based on the learnt descriptor to improve recognition performance. Experimental results on standard WSLR datasets show that our method outperforms previous state-of-the-art methods significantly. We also demonstrate the effectiveness of our method on automatically localizing signs from sign news, achieving 28.1 for AP@0.5.

cs.CV↗

arXiv2020

Word-level Deep Sign Language Recognition from Video: A New Large-scale Dataset and Methods Comparison

Vision-based sign language recognition aims at helping deaf people to communicate with others. However, most existing sign language datasets are limited to a small number of words. Due to the limited vocabulary size, models learned from those datasets cannot be applied in practice. In this paper, we introduce a new large-scale Word-Level American Sign Language (WLASL) video dataset, containing more than 2000 words performed by over 100 signers. This dataset will be made publicly available to the research community. To our knowledge, it is by far the largest public ASL dataset to facilitate word-level sign recognition research. Based on this new large-scale dataset, we are able to experiment with several deep learning methods for word-level sign recognition and evaluate their performances in large scale scenarios. Specifically we implement and compare two different models,i.e., (i) holistic visual appearance-based approach, and (ii) 2D human pose based approach. Both models are valuable baselines that will benefit the community for method benchmarking. Moreover, we also propose a novel pose-based temporal graph convolution networks (Pose-TGCN) that models spatial and temporal dependencies in human pose trajectories simultaneously, which has further boosted the performance of the pose-based method. Our results show that pose-based and appearance-based models achieve comparable performances up to 66% at top-10 accuracy on 2,000 words/glosses, demonstrating the validity and challenges of our dataset. Our dataset and baseline deep models are available at \url{https://dxli94.github.io/WLASL/}.

cs.CV↗

arXiv2016

Controllable deposition of titanium dioxides onto carbon nanotubes in aqueous solutions

Within the field of nanotechnology, nano-scale composites have significant potential in the development of advanced materials for functional applications. Here, composites based on carbon nanotubes and titanium dioxides have been prepared with titanyl sulfate using a chemical bath deposition method at or near room temperature. Two kinds of titanium dioxide depositions corresponding to rutile and anatase were sheathed on carbon nanotubes evenly by adjusting the precursor concentration and temperature. Possible composite mechanisms are discussed. Compared with original tubes, the specific surface areas have been improved nearly four times and the content ratio of mesostructures has also been increased after deposition processes.

cond-mat.mtrl-sci↗

arXiv2015

Nitrogen oxides under pressure stability, ionization, polymerization, and superconductivity

Nitrogen oxides are textbook class of molecular compounds, with extensive industrial applications. Nitrogen and oxygen are also among the most abundant elements in the universe. We explore the N-O system at 0 K and up to 500 GPa though ab initio evolutionary simulations. Results show that two phase transformations of stable molecular NO2 exist at 7 and 64 GPa, and followed by decomposition of NO2 at 91 GPa. All of the NO+NO3- structures are found to be metastable at T=0 K, so experimentally reported ionic NO+NO3- is either metastable or stabilized by temperature. Upon increasing pressure, N2O5 transforms from P-1 to C2/c structure at 51 GPa. NO becomes thermodynamically stable at 198 GPa. This polymeric phase is superconducting (Tc = 2.0 K) and contains a -N-N- backbone.

cond-mat.mtrl-sci↗

arXiv2014

Object Selection under Team Context

Context-aware database has drawn increasing attention from both industry and academia recently by taking users' current situation and environment into consideration. However, most of the literature focus on individual context, overlooking the team users. In this paper, we investigate how to integrate team context into database query process to help the users' get top-ranked database tuples and make the team more competitive. We introduce naive and optimized query algorithm to select the suitable records and show that they output the same results while the latter is more computational efficient. Extensive empirical studies are conducted to evaluate the query approaches and demonstrate their effectiveness and efficiency.

cs.DB↗

arXiv2014

A 2D based Partition Strategy for Solving Ranking under Team Context (RTP)

In this paper, we propose a 2D based partition method for solving the problem of Ranking under Team Context(RTC) on datasets without a priori. We first map the data into 2D space using its minimum and maximum value among all dimensions. Then we construct window queries with consideration of current team context. Besides, during the query mapping procedure, we can pre-prune some tuples which are not top ranked ones. This pre-classified step will defer processing those tuples and can save cost while providing solutions for the problem. Experiments show that our algorithm performs well especially on large datasets with correctness.

cs.DB↗