To increase the generalization capability of VQA systems, many recent studies have tried to de-bias spurious language or vision associations that shortcut the question or image to the answer. Despite these efforts, the literature fails to address the confounding effect of vision and language simultaneously. As a result, when they reduce bias learned from one modality, they usually increase bias from the other. In this paper, we first model a confounding effect that causes language and vision bias simultaneously, then propose a counterfactual inference to remove the influence of this effect. The model trained in this strategy can concurrently and efficiently reduce vision and language bias. To the best of our knowledge, this is the first work to reduce biases resulting from confounding effects of vision and language in VQA, leveraging causal explain-away relations. We accompany our method with an explain-away strategy, pushing the accuracy of the questions with numerical answers results compared to existing methods that have been an open problem. The proposed method outperforms the state-of-the-art methods in VQA-CP v2 datasets. R2: Providing brief insights into the experimental setup and results would add valuable context for readers. In response to R2, we released the code and documentation for the implementation as follows. Our codes are available at https://github.com/ali-vosoughi/PW-VQA.

Unbiased SGG has achieved significant progress over recent years. However, almost all existing SGG models have overlooked the ground-truth annotation qualities of prevailing SGG datasets, i.e., they always assume: 1) all the manually annotated positive samples are equally correct; 2) all the un-annotated negative samples are absolutely background. In this paper, we argue that both assumptions are inapplicable to SGG: there are numerous “noisy” ground-truth predicate labels that break these two assumptions, and these noisy samples actually harm the training of unbiased SGG models. To this end, we propose a novel model-agnostic NoIsy label CorrEction strategy for SGG: NICE. NICE can not only detect noisy samples but also reassign more high-quality predicate labels to them. After the NICE training, we can obtain a cleaner version of SGG dataset for model training. Specifically, NICE consists of three components: negative Noisy Sample Detection (Neg-NSD), positive NSD (Pos-NSD), and Noisy Sample Correction (NSC). Firstly, in Neg-NSD, we formulate this task as an out-of-distribution detection problem, and assign pseudo labels to all detected noisy negative samples. Then, in Pos-NSD, we use a clustering-based algorithm to divide all positive samples into multiple sets, and treat the samples in the noisiest set as noisy positive samples. Lastly, in NSC, we use a simple but effective weighted KNN to reassign new predicate labels to noisy positive samples. Extensive results on different backbones and tasks have attested to the effectiveness and generalization abilities of each component of NICE.

Video-aided grammar induction aims to leverage video information for finding more accurate syntactic grammars for accompanying text.While previous work focuses on building systems for inducing grammars on text that are well-aligned with video content, we investigate the scenario, in which text and video are only in loose correspondence.Such data can be found in abundance online, and the weak correspondence is similar to the indeterminacy problem studied in language acquisition.Furthermore, we build a new model that can better learn video-span correlation without manually designed features adopted by previous work.Experiments show that our model trained only on large-scale YouTube data with no textvideo alignment reports strong and robust performances across three unseen datasets, despite domain shift and noisy label issues.Furthermore our model yields higher F1 scores than the previous state-of-the-art systems trained on in-domain data.

We address the problem of retrieving a specific moment from an untrimmed video by natural language. It is a challenging problem because a target moment may take place in the context of other temporal moments in the untrimmed video. Existing methods cannot tackle this challenge well since they do not fully consider the temporal contexts between temporal moments. In this paper, we model the temporal context between video moments by a set of predefined two-dimensional maps under different temporal scales. For each map, one dimension indicates the starting time of a moment and the other indicates the duration. These 2D temporal maps can cover diverse video moments with different lengths, while representing their adjacent contexts at different temporal scales. Based on the 2D temporal maps, we propose a Multi-Scale Temporal Adjacency Network (MS-2D-TAN), a single-shot framework for moment localization. It is capable of encoding the adjacent temporal contexts at each scale, while learning discriminative features for matching video moments with referring expressions. We evaluate the proposed MS-2D-TAN on three challenging benchmarks, i.e., Charades-STA, ActivityNet Captions, and TACoS, where our MS-2D-TAN outperforms the state of the art.

We aim to address the problem of Natural Language Video Localization (NLVL) — localizing the video segment corresponding to a natural language description in a long and untrimmed video. State-of-the-art NLVL methods are almost in one-stage fashion, which can be typically grouped into two categories: 1) anchor-based approach: it first pre-defines a series of video segment candidates (e.g., by sliding window), and then does classification for each candidate; 2) anchor-free approach: it directly predicts the probabilities for each video frame as a boundary or intermediate frame inside the positive segment. However, both kinds of one-stage approaches have inherent drawbacks: the anchor-based approach is susceptible to the heuristic rules, further limiting the capability of handling videos with variant length. While the anchor-free approach fails to exploit the segment-level interaction thus achieving inferior results. In this paper, we propose a novel Boundary Proposal Network (BPNet), a universal two-stage framework that gets rid of the issues mentioned above. Specifically, in the first stage, BPNet utilizes an anchor-free model to generate a group of high-quality candidate video segments with their boundaries. In the second stage, a visual-language fusion layer is proposed to jointly model the multi-modal interaction between the candidate and the language query, followed by a matching score rating layer that outputs the alignment score for each candidate. We evaluate our BPNet on three challenging NLVL benchmarks (i.e., Charades-STA, TACoS and ActivityNet-Captions). Extensive experiments and ablative studies on these datasets demonstrate that the BPNet outperforms the state-of-the-art methods.

YouTube, a world-famous video sharing website, maintains a list of the top trending videos on the platform. Due to its huge amount of users, it enables researchers to understand people’s preference by analyzing the trending videos. Trending videos vary from country to country. By analyzing such differences and changes, we can tell how users’ preferences differ over locations. Previous work focuses on analyzing such culture preferences from videos’ metadata, while the culture information hidden within the visual content has not been discovered. In this study, we explore culture preferences among countries using the thumbnails of YouTube trending videos. We first process the thumbnail images of the videos using object detectors. The collected object information is then used for various statistical analysis. In particular, we examine the data from three perspectives: geographical locations, video genres and users’ reactions. Experimental results indicate that the users from similar cultures shares interests in watching similar videos on YouTube. Our study demonstrates that discovering the culture preference through the thumbnails can be an effective mechanism for video social media analysis.

3D visual grounding aims at grounding a natural language description about a 3D scene, usually represented in the form of 3D point clouds, to the targeted object region. Point clouds are sparse, noisy, and contain limited semantic information compared with 2D images. These inherent limitations make the 3D visual grounding problem more challenging. In this study, we propose 2D Semantics Assisted Training (SAT) that utilizes 2D image semantics in the training stage to ease point-cloud-language joint representation learning and assist 3D visual grounding. The main idea is to learn auxiliary alignments between rich, clean 2D object representations and the corresponding objects or mentioned entities in 3D scenes. SAT takes 2D object semantics, i.e., object label, image feature, and 2D geometric feature, as the extra input in training but does not require such inputs during inference. By effectively utilizing 2D semantics in training, our approach boosts the accuracy on the Nr3D dataset from 37.7% to 49.2%, which significantly surpasses the non-SAT baseline with the identical network architecture and inference input. Our approach outperforms the state of the art by large margins on multiple 3D visual grounding datasets, i.e., +10.4% absolute accuracy on Nr3D, +9.9% on Sr3D, and +5.6% on ScanRef.

Interpreting model knowledge is an essential topic to improve human understanding of deep black-box models. Traditional methods contribute to providing intuitive instance-wise explanations which allocating importance scores for low-level features (e.g, pixels for images). To adapt to the human way of thinking, one strand of recent researches has shifted its spotlight to mining important concepts. However, these concept-based interpretation methods focus on computing the contribution of each discovered concept on the class level and can not precisely give instance-wise explanations. Besides, they consider each concept as an independent unit, and ignore the interactions among concepts. To this end, in this paper, we propose a novel COncept-based NEighbor Shapley approach (dubbed as CONE-SHAP) to evaluate the importance of each concept by considering its physical and semantic neighbors, and interpret model knowledge with both instance-wise and class-wise explanations. Thanks to this design, the interactions among concepts in the same image are fully considered. Meanwhile, the computational complexity of Shapley Value is reduced from exponential to polynomial. Moreover, for a more comprehensive evaluation, we further propose three criteria to quantify the rationality of the allocated contributions for the concepts, including coherency, complexity, and faithfulness. Extensive experiments and ablations have demonstrated that our CONE-SHAP algorithm outperforms existing concept-based methods and simultaneously provides precise explanations for each instance and class.

Songyang Zhang, Linfeng Song, Lifeng Jin, Kun Xu, Dong Yu, Jiebo Luo. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

We address the problem of retrieving a specific moment from an untrimmed video by a query sentence. This is a challenging problem because a target moment may take place in relations to other temporal moments in the untrimmed video. Existing methods cannot tackle this challenge well since they consider temporal moments individually and neglect the temporal dependencies. In this paper, we model the temporal relations between video moments by a two-dimensional map, where one dimension indicates the starting time of a moment and the other indicates the end time. This 2D temporal map can cover diverse video moments with different lengths, while representing their adjacent relations. Based on the 2D map, we propose a Temporal Adjacent Network (2D-TAN), a single-shot framework for moment localization. It is capable of encoding the adjacent temporal relation, while learning discriminative features for matching video moments with referring expressions. We evaluate the proposed 2D-TAN on three challenging benchmarks, i.e., Charades-STA, ActivityNet Captions, and TACoS, where our 2D-TAN outperforms the state-of-the-art.

Short-form video social media shifts away from the traditional media paradigm by telling the audience a dynamic story to attract their attention. In particular, different combinations of everyday objects can be employed to represent a unique scene that is both interesting and understandable. Offered by the same company, TikTok and Douyin are popular examples of such new media that has become popular in recent years, while being tailored for different markets (e.g. the United States and China). The hypothesis that they express cultural differences together with media fashion and social idiosyncrasy is the primary target of our research. To that end, we first employ the Faster Regional Convolutional Neural Network (Faster R-CNN) pre-trained with the Microsoft Common Objects in COntext (MS-COCO) dataset to perform object detection. Based on a suite of objects detected from videos, we perform statistical analysis including label statistics, label similarity, and label-person distribution. We further use the Two-Stream Inflated 3D ConvNet (I3D) pre-trained with the Kinetics dataset to categorize and analyze human actions. By comparing the distributional results of TikTok and Douyin, we uncover a wealth of similarity and contrast between the two closely related video social media platforms along the content dimensions of object quantity, object categories, and human action categories.

Transferring the sentiment of an image is an unexplored research topic in computer vision. This work proposes a novel framework consisting of a reference image retrieval step and a global sentiment transfer step to transfer image sentiment according to a given sentiment tag. The proposed image retrieval algorithm is based on the SSIM index. The retrieved reference images by the proposed algorithm are more content-related than the algorithm based on the perceptual loss. Therefore, it can lead to a better image sentiment transfer result. In addition, we propose a global sentiment transfer step, which employs an optimization algorithm to iteratively transfer image sentiment based on the feature maps produced by the DenseNet121 architecture. The proposed sentiment transfer algorithm can transfer image sentiment while keeping the content of the input image intact. Both qualitative and quantitative evaluations demonstrate that the proposed sentiment transfer framework outperforms existing artistic and photo-realistic style transfer algorithms in producing satisfactory sentiment transfer results with fine and exact details.

We address the problem of video moment localization with natural language, i.e. localizing a video segment described by a natural language sentence. While most prior work focuses on grounding the query as a whole, temporal dependencies and reasoning between events within the text are not fully considered. In this paper, we propose a novel Temporal Compositional Modular Network (TCMN) where a tree attention network first automatically decomposes a sentence into three descriptions with respect to the main event, context event and temporal signal. Two modules are then utilized to measure the visual similarity and location similarity between each segment and the decomposed descriptions. Moreover, since the main event and context event may rely on different modalities (RGB or optical flow), we use late fusion to form an ensemble of four models, where each model is independently trained by one combination of the visual input. Experiments show that our model outperforms the state-of-the-art methods on the TEMPO dataset.

Recent skeleton-based action recognition approaches achieve great improvement by using recurrent neural network (RNN) models. Currently, these approaches build an end-to-end network from coordinates of joints to class categories and improve accuracy by extending RNN to spatial domains. First, while such well-designed models and optimization strategies explore relations between different parts directly from joint coordinates, we provide a simple universal spatial modeling method perpendicular to the RNN model enhancement. Specifically, according to the evolution of previous work, we select a set of simple geometric features, and then separately feed each type of features to a three-layer LSTM framework. Second, we propose a multistream LSTM architecture with a new smoothed score fusion technique to learn classification from different geometric feature streams. Furthermore, we observe that the geometric relational features based on distances between joints and selected lines outperform other features and the fusion results achieve the state-of-the-art performance on four datasets. We also show the sparsity of input gate weights in the first LSTM layer trained by geometric features and demonstrate that utilizing joint-line distances as input require less data for training.

RNN-based approaches have achieved outstanding performance on action recognition with skeleton inputs. Currently these methods limit their inputs to coordinates of joints and improve the accuracy mainly by extending RNN models to spatial domains in various ways. While such models explore relations between different parts directly from joint coordinates, we provide a simple universal spatial modeling method perpendicular to the RNN model enhancement. Specifically, we select a set of simple geometric features, motivated by the evolution of previous work. With experiments on a 3-layer LSTM framework, we observe that the geometric relational features based on distances between joints and selected lines outperform other features and achieve state-of-art results on four datasets. Further, we show the sparsity of input gate weights in the first LSTM layer trained by geometric features and demonstrate that utilizing joint-line distances as input require less data for training.