Shin'ichi Satoh

Hui Wei, Zhixiang Wang, Xuemei Jia et al.

Adversarial attacks on thermal infrared imaging expose the risk of related applications. Estimating the security of these systems is essential for safely deploying them in the real world. In many cases, realizing the attacks in the physical space requires elaborate special perturbations. These solutions are often \emph{impractical} and \emph{attention-grabbing}. To address the need for a physically practical and stealthy adversarial attack, we introduce \textsc{HotCold} Block, a novel physical attack for infrared detectors that hide persons utilizing the wearable Warming Paste and Cooling Paste. By attaching these readily available temperature-controlled materials to the body, \textsc{HotCold} Block evades human eyes efficiently. Moreover, unlike existing methods that build adversarial patches with complex texture and structure features, \textsc{HotCold} Block utilizes an SSP-oriented adversarial optimization algorithm that enables attacks with pure color blocks and explores the influence of size, shape, and position on attack performance. Extensive experimental results in both digital and physical environments demonstrate the performance of our proposed \textsc{HotCold} Block. \emph{Code is available: \textcolor{magenta}{https://github.com/weihui1308/HOTCOLDBlock}}.

Liang Liao, Wenyi Chen, Jing Xiao et al.

Understanding foggy image sequence in the driving scenes is critical for autonomous driving, but it remains a challenging task due to the difficulty in collecting and annotating real-world images of adverse weather. Recently, the self-training strategy has been considered a powerful solution for unsupervised domain adaptation, which iteratively adapts the model from the source domain to the target domain by generating target pseudo labels and re-training the model. However, the selection of confident pseudo labels inevitably suffers from the conflict between sparsity and accuracy, both of which will lead to suboptimal models. To tackle this problem, we exploit the characteristics of the foggy image sequence of driving scenes to densify the confident pseudo labels. Specifically, based on the two discoveries of local spatial similarity and adjacent temporal correspondence of the sequential image data, we propose a novel Target-Domain driven pseudo label Diffusion (TDo-Dif) scheme. It employs superpixels and optical flows to identify the spatial similarity and temporal correspondence, respectively and then diffuses the confident but sparse pseudo labels within a superpixel or a temporal corresponding pair linked by the flow. Moreover, to ensure the feature similarity of the diffused pixels, we introduce local spatial similarity loss and temporal contrastive loss in the model re-training stage. Experimental results show that our TDo-Dif scheme helps the adaptive model achieve 51.92% and 53.84% mean intersection-over-union (mIoU) on two publicly available natural foggy datasets (Foggy Zurich and Foggy Driving), which exceeds the state-of-the-art unsupervised domain adaptive semantic segmentation methods. Models and data can be found at https://github.com/velor2012/TDo-Dif.

Kejun Lin, Zhixiang Wang, Zheng Wang et al.

Person re-identification (re-ID) requires densely distributed cameras. In practice, the person of interest may not be captured by cameras and, therefore, needs to be retrieved using subjective information (e.g., sketches from witnesses). Previous research defines this case using the sketch as sketch re-identification (Sketch re-ID) and focuses on eliminating the domain gap. Actually, subjectivity is another significant challenge. We model and investigate it by posing a new dataset with multi-witness descriptions. It features two aspects. 1) Large-scale. It contains over 4,763 sketches and 32,668 photos, making it the largest Sketch re-ID dataset. 2) Multi-perspective and multi-style. Our dataset offers multiple sketches for each identity. Witnesses' subjective cognition provides multiple perspectives on the same individual, while different artists' drawing styles provide variation in sketch styles. We further have two novel designs to alleviate the challenge of subjectivity. 1) Fusing subjectivity. We propose a non-local (NL) fusion module that gathers sketches from different witnesses for the same identity. 2) Introducing objectivity. An AttrAlign module utilizes attributes as an implicit mask to align cross-domain features. To push forward the advance of Sketch re-ID, we set three benchmarks (large-scale, multi-style, cross-style). Extensive experiments demonstrate our leading performance in these benchmarks. Dataset and Codes are publicly available at: https://github.com/Lin-Kayla/subjectivity-sketch-reid

Taorong Liu, Liang Liao, Delin Chen et al.

Image inpainting for completing complicated semantic environments and diverse hole patterns of corrupted images is challenging even for state-of-the-art learning-based inpainting methods trained on large-scale data. A reference image capturing the same scene of a corrupted image offers informative guidance for completing the corrupted image as it shares similar texture and structure priors to that of the holes of the corrupted image. In this work, we propose a transformer-based encoder-decoder network, named TransRef, for reference-guided image inpainting. Specifically, the guidance is conducted progressively through a reference embedding procedure, in which the referencing features are subsequently aligned and fused with the features of the corrupted image. For precise utilization of the reference features for guidance, a reference-patch alignment (Ref-PA) module is proposed to align the patch features of the reference and corrupted images and harmonize their style differences, while a reference-patch transformer (Ref-PT) module is proposed to refine the embedded reference feature. Moreover, to facilitate the research of reference-guided image restoration tasks, we construct a publicly accessible benchmark dataset containing 50K pairs of input and reference images. Both quantitative and qualitative evaluations demonstrate the efficacy of the reference information and the proposed method over the state-of-the-art methods in completing complex holes. Code and dataset can be accessed at https://github.com/Cameltr/TransRef.

Hui Wei, Hao Tang, Xuemei Jia et al.

Despite the impressive achievements of Deep Neural Networks (DNNs) in computer vision, their vulnerability to adversarial attacks remains a critical concern. Extensive research has demonstrated that incorporating sophisticated perturbations into input images can lead to a catastrophic degradation in DNNs' performance. This perplexing phenomenon not only exists in the digital space but also in the physical world. Consequently, it becomes imperative to evaluate the security of DNNs-based systems to ensure their safe deployment in real-world scenarios, particularly in security-sensitive applications. To facilitate a profound understanding of this topic, this paper presents a comprehensive overview of physical adversarial attacks. Firstly, we distill four general steps for launching physical adversarial attacks. Building upon this foundation, we uncover the pervasive role of artifacts carrying adversarial perturbations in the physical world. These artifacts influence each step. To denote them, we introduce a new term: adversarial medium. Then, we take the first step to systematically evaluate the performance of physical adversarial attacks, taking the adversarial medium as a first attempt. Our proposed evaluation metric, hiPAA, comprises six perspectives: Effectiveness, Stealthiness, Robustness, Practicability, Aesthetics, and Economics. We also provide comparative results across task categories, together with insightful observations and suggestions for future research directions.

Zhixiang Wang, Xiang Ji, Jia-Bin Huang et al.

Most computer vision systems assume distortion-free images as inputs. The widely used rolling-shutter (RS) image sensors, however, suffer from geometric distortion when the camera and object undergo motion during capture. Extensive researches have been conducted on correcting RS distortions. However, most of the existing work relies heavily on the prior assumptions of scenes or motions. Besides, the motion estimation steps are either oversimplified or computationally inefficient due to the heavy flow warping, limiting their applicability. In this paper, we investigate using rolling shutter with a global reset feature (RSGR) to restore clean global shutter (GS) videos. This feature enables us to turn the rectification problem into a deblur-like one, getting rid of inaccurate and costly explicit motion estimation. First, we build an optic system that captures paired RSGR/GS videos. Second, we develop a novel algorithm incorporating spatial and temporal designs to correct the spatial-varying RSGR distortion. Third, we demonstrate that existing image-to-image translation algorithms can recover clean GS videos from distorted RSGR inputs, yet our algorithm achieves the best performance with the specific designs. Our rendered results are not only visually appealing but also beneficial to downstream tasks. Compared to the state-of-the-art RS solution, our RSGR solution is superior in both effectiveness and efficiency. Considering it is easy to realize without changing the hardware, we believe our RSGR solution can potentially replace the RS solution in taking distortion-free videos with low noise and low budget.

Zelong Zeng, Zheng Wang, Fan Yang et al.

The large variation of viewpoint and irrelevant content around the target always hinder accurate image retrieval and its subsequent tasks. In this paper, we investigate an extremely challenging task: given a ground-view image of a landmark, we aim to achieve cross-view geo-localization by searching out its corresponding satellite-view images. Specifically, the challenge comes from the gap between ground-view and satellite-view, which includes not only large viewpoint changes (some parts of the landmark may be invisible from front view to top view) but also highly irrelevant background (the target landmark tend to be hidden in other surrounding buildings), making it difficult to learn a common representation or a suitable mapping. To address this issue, we take advantage of drone-view information as a bridge between ground-view and satellite-view domains. We propose a Peer Learning and Cross Diffusion (PLCD) framework. PLCD consists of three parts: 1) a peer learning across ground-view and drone-view to find visible parts to benefit ground-drone cross-view representation learning; 2) a patch-based network for satellite-drone cross-view representation learning; 3) a cross diffusion between ground-drone space and satellite-drone space. Extensive experiments conducted on the University-Earth and University-Google datasets show that our method outperforms state-of-the-arts significantly.

Mayu Otani, Riku Togashi, Yu Sawai et al.

Human evaluation is critical for validating the performance of text-to-image generative models, as this highly cognitive process requires deep comprehension of text and images. However, our survey of 37 recent papers reveals that many works rely solely on automatic measures (e.g., FID) or perform poorly described human evaluations that are not reliable or repeatable. This paper proposes a standardized and well-defined human evaluation protocol to facilitate verifiable and reproducible human evaluation in future works. In our pilot data collection, we experimentally show that the current automatic measures are incompatible with human perception in evaluating the performance of the text-to-image generation results. Furthermore, we provide insights for designing human evaluation experiments reliably and conclusively. Finally, we make several resources publicly available to the community to facilitate easy and fast implementations.

Taorong Liu, Liang Liao, Zheng Wang et al.

Existing learning-based image inpainting methods are still in challenge when facing complex semantic environments and diverse hole patterns. The prior information learned from the large scale training data is still insufficient for these situations. Reference images captured covering the same scenes share similar texture and structure priors with the corrupted images, which offers new prospects for the image inpainting tasks. Inspired by this, we first build a benchmark dataset containing 10K pairs of input and reference images for reference-guided inpainting. Then we adopt an encoder-decoder structure to separately infer the texture and structure features of the input image considering their pattern discrepancy of texture and structure during inpainting. A feature alignment module is further designed to refine these features of the input image with the guidance of a reference image. Both quantitative and qualitative evaluations demonstrate the superiority of our method over the state-of-the-art methods in terms of completing complex holes.

Mayu Otani, Riku Togashi, Yuta Nakashima et al.

Mean Average Precision (mAP) is the primary evaluation measure for object detection. Although object detection has a broad range of applications, mAP evaluates detectors in terms of the performance of ranked instance retrieval. Such the assumption for the evaluation task does not suit some downstream tasks. To alleviate the gap between downstream tasks and the evaluation scenario, we propose Optimal Correction Cost (OC-cost), which assesses detection accuracy at image level. OC-cost computes the cost of correcting detections to ground truths as a measure of accuracy. The cost is obtained by solving an optimal transportation problem between the detections and the ground truths. Unlike mAP, OC-cost is designed to penalize false positive and false negative detections properly, and every image in a dataset is treated equally. Our experimental result validates that OC-cost has better agreement with human preference than a ranking-based measure, i.e., mAP for a single image. We also show that detectors' rankings by OC-cost are more consistent on different data splits than mAP. Our goal is not to replace mAP with OC-cost but provide an additional tool to evaluate detectors from another aspect. To help future researchers and developers choose a target measure, we provide a series of experiments to clarify how mAP and OC-cost differ.

Zelong Zeng, Fan Yang, Hong Liu et al.

Recent deep metric learning (DML) methods typically leverage solely class labels to keep positive samples far away from negative ones. However, this type of method normally ignores the crucial knowledge hidden in the data (e.g., intra-class information variation), which is harmful to the generalization of the trained model. To alleviate this problem, in this paper we propose Online Batch Diffusion-based Self-Distillation (OBD-SD) for DML. Specifically, we first propose a simple but effective Progressive Self-Distillation (PSD), which distills the knowledge progressively from the model itself during training. The soft distance targets achieved by PSD can present richer relational information among samples, which is beneficial for the diversity of embedding representations. Then, we extend PSD with an Online Batch Diffusion Process (OBDP), which is to capture the local geometric structure of manifolds in each batch, so that it can reveal the intrinsic relationships among samples in the batch and produce better soft distance targets. Note that our OBDP is able to restore the insufficient manifold relationships obtained by the original PSD and achieve significant performance improvement. Our OBD-SD is a flexible framework that can be integrated into state-of-the-art (SOTA) DML methods. Extensive experiments on various benchmarks, namely CUB200, CARS196, and Stanford Online Products, demonstrate that our OBD-SD consistently improves the performance of the existing DML methods on multiple datasets with negligible additional training time, achieving very competitive results. Code: \url{https://github.com/ZelongZeng/OBD-SD_Pytorch}

Astha Verma, A V Subramanyam, Siddhesh Bangar et al.

Certified defense methods against adversarial perturbations have been recently investigated in the black-box setting with a zeroth-order (ZO) perspective. However, these methods suffer from high model variance with low performance on high-dimensional datasets due to the ineffective design of the denoiser and are limited in their utilization of ZO techniques. To this end, we propose a certified ZO preprocessing technique for removing adversarial perturbations from the attacked image in the black-box setting using only model queries. We propose a robust UNet denoiser (RDUNet) that ensures the robustness of black-box models trained on high-dimensional datasets. We propose a novel black-box denoised smoothing (DS) defense mechanism, ZO-RUDS, by prepending our RDUNet to the black-box model, ensuring black-box defense. We further propose ZO-AE-RUDS in which RDUNet followed by autoencoder (AE) is prepended to the black-box model. We perform extensive experiments on four classification datasets, CIFAR-10, CIFAR-10, Tiny Imagenet, STL-10, and the MNIST dataset for image reconstruction tasks. Our proposed defense methods ZO-RUDS and ZO-AE-RUDS beat SOTA with a huge margin of $35\%$ and $9\%$, for low dimensional (CIFAR-10) and with a margin of $20.61\%$ and $23.51\%$ for high-dimensional (STL-10) datasets, respectively.

Zhixiang Wang, Yu-Lun Liu, Jia-Bin Huang et al.

Close-up facial images captured at short distances often suffer from perspective distortion, resulting in exaggerated facial features and unnatural/unattractive appearances. We propose a simple yet effective method for correcting perspective distortions in a single close-up face. We first perform GAN inversion using a perspective-distorted input facial image by jointly optimizing the camera intrinsic/extrinsic parameters and face latent code. To address the ambiguity of joint optimization, we develop starting from a short distance, optimization scheduling, reparametrizations, and geometric regularization. Re-rendering the portrait at a proper focal length and camera distance effectively corrects perspective distortions and produces more natural-looking results. Our experiments show that our method compares favorably against previous approaches qualitatively and quantitatively. We showcase numerous examples validating the applicability of our method on in-the-wild portrait photos. We will release our code and the evaluation protocol to facilitate future work.

Zelong Zeng, Fan Yang, Zheng Wang et al.

Most deep metric learning (DML) methods employ a strategy that forces all positive samples to be close in the embedding space while keeping them away from negative ones. However, such a strategy ignores the internal relationships of positive (negative) samples and often leads to overfitting, especially in the presence of hard samples and mislabeled samples. In this work, we propose a simple yet effective regularization, namely Listwise Self-Distillation (LSD), which progressively distills a model's own knowledge to adaptively assign a more appropriate distance target to each sample pair in a batch. LSD encourages smoother embeddings and information mining within positive (negative) samples as a way to mitigate overfitting and thus improve generalization. Our LSD can be directly integrated into general DML frameworks. Extensive experiments show that LSD consistently boosts the performance of various metric learning methods on multiple datasets.

Andreu Girbau, Ferran Marqués, Shin'ichi Satoh

Current approaches in Multiple Object Tracking (MOT) rely on the spatio-temporal coherence between detections combined with object appearance to match objects from consecutive frames. In this work, we explore MOT using object appearances as the main source of association between objects in a video, using spatial and temporal priors as weighting factors. We form initial tracklets by leveraging on the idea that instances of an object that are close in time should be similar in appearance, and build the final object tracks by fusing the tracklets in a hierarchical fashion. We conduct extensive experiments that show the effectiveness of our method over three different MOT benchmarks, MOT17, MOT20, and DanceTrack, being competitive in MOT17 and MOT20 and establishing state-of-the-art results in DanceTrack.

Zhixiang Wang, Baiang Li, Jian Wang et al.

This paper introduces an innovative approach for image matting that redefines the traditional regression-based task as a generative modeling challenge. Our method harnesses the capabilities of latent diffusion models, enriched with extensive pre-trained knowledge, to regularize the matting process. We present novel architectural innovations that empower our model to produce mattes with superior resolution and detail. The proposed method is versatile and can perform both guidance-free and guidance-based image matting, accommodating a variety of additional cues. Our comprehensive evaluation across three benchmark datasets demonstrates the superior performance of our approach, both quantitatively and qualitatively. The results not only reflect our method's robust effectiveness but also highlight its ability to generate visually compelling mattes that approach photorealistic quality. The project page for this paper is available at https://lightchaserx.github.io/matting-by-generation/

Andreu Girbau-Xalabarder, Jun Nagata, Shinichi Sumiyoshi et al.

Event cameras capture scene changes asynchronously on a per-pixel basis, enabling extremely high temporal resolution. However, this advantage comes at the cost of high bandwidth, memory, and computational demands. To address this, prior work has explored event downsampling, but most approaches rely on fixed heuristics or threshold-based strategies, limiting their adaptability. Instead, we propose a probabilistic framework, POLED, that models event importance through an event-importance probability density function (ePDF), which can be arbitrarily defined and adapted to different applications. Our approach operates in a purely online setting, estimating event importance on-the-fly from raw event streams, enabling scene-specific adaptation. Additionally, we introduce zero-shot event downsampling, where downsampled events must remain usable for models trained on the original event stream, without task-specific adaptation. We design a contour-preserving ePDF that prioritizes structurally important events and evaluate our method across four datasets and tasks--object classification, image interpolation, surface normal estimation, and object detection--demonstrating that intelligent sampling is crucial for maintaining performance under event-budget constraints. Code available.

Zheng Wang, Zhixiang Wang, Yinqiang Zheng et al.

An efficient and effective person re-identification (ReID) system relieves the users from painful and boring video watching and accelerates the process of video analysis. Recently, with the explosive demands of practical applications, a lot of research efforts have been dedicated to heterogeneous person re-identification (Hetero-ReID). In this paper, we provide a comprehensive review of state-of-the-art Hetero-ReID methods that address the challenge of inter-modality discrepancies. According to the application scenario, we classify the methods into four categories -- low-resolution, infrared, sketch, and text. We begin with an introduction of ReID, and make a comparison between Homogeneous ReID (Homo-ReID) and Hetero-ReID tasks. Then, we describe and compare existing datasets for performing evaluations, and survey the models that have been widely employed in Hetero-ReID. We also summarize and compare the representative approaches from two perspectives, i.e., the application scenario and the learning pipeline. We conclude by a discussion of some future research directions. Follow-up updates are avaible at: https://github.com/lightChaserX/Awesome-Hetero-reID

Yusuke Uchida, Yuki Nagai, Shigeyuki Sakazawa et al.

Deep neural networks have recently achieved significant progress. Sharing trained models of these deep neural networks is very important in the rapid progress of researching or developing deep neural network systems. At the same time, it is necessary to protect the rights of shared trained models. To this end, we propose to use a digital watermarking technology to protect intellectual property or detect intellectual property infringement of trained models. Firstly, we formulate a new problem: embedding watermarks into deep neural networks. We also define requirements, embedding situations, and attack types for watermarking to deep neural networks. Secondly, we propose a general framework to embed a watermark into model parameters using a parameter regularizer. Our approach does not hurt the performance of networks into which a watermark is embedded. Finally, we perform comprehensive experiments to reveal the potential of watermarking to deep neural networks as a basis of this new problem. We show that our framework can embed a watermark in the situations of training a network from scratch, fine-tuning, and distilling without hurting the performance of a deep neural network. The embedded watermark does not disappear even after fine-tuning or parameter pruning; the watermark completely remains even after removing 65% of parameters were pruned. The implementation of this research is: https://github.com/yu4u/dnn-watermark

Zhijing Wan, Zhixiang Wang, Yuran Wang et al.

Coreset selection seeks to choose a subset of crucial training samples for efficient learning. It has gained traction in deep learning, particularly with the surge in training dataset sizes. Sample selection hinges on two main aspects: a sample's representation in enhancing performance and the role of sample diversity in averting overfitting. Existing methods typically measure both the representation and diversity of data based on similarity metrics, such as L2-norm. They have capably tackled representation via distribution matching guided by the similarities of features, gradients, or other information between data. However, the results of effectively diverse sample selection are mired in sub-optimality. This is because the similarity metrics usually simply aggregate dimension similarities without acknowledging disparities among the dimensions that significantly contribute to the final similarity. As a result, they fall short of adequately capturing diversity. To address this, we propose a feature-based diversity constraint, compelling the chosen subset to exhibit maximum diversity. Our key lies in the introduction of a novel Contributing Dimension Structure (CDS) metric. Different from similarity metrics that measure the overall similarity of high-dimensional features, our CDS metric considers not only the reduction of redundancy in feature dimensions, but also the difference between dimensions that contribute significantly to the final similarity. We reveal that existing methods tend to favor samples with similar CDS, leading to a reduced variety of CDS types within the coreset and subsequently hindering model performance. In response, we enhance the performance of five classical selection methods by integrating the CDS constraint. Our experiments on three datasets demonstrate the general effectiveness of the proposed method in boosting existing methods.

Avinash Anand, Raj Jaiswal, Pijush Bhuyan et al.

The automatic recognition of tabular data in document images presents a significant challenge due to the diverse range of table styles and complex structures. Tables offer valuable content representation, enhancing the predictive capabilities of various systems such as search engines and Knowledge Graphs. Addressing the two main problems, namely table detection (TD) and table structure recognition (TSR), has traditionally been approached independently. In this research, we propose an end-to-end pipeline that integrates deep learning models, including DETR, CascadeTabNet, and PP OCR v2, to achieve comprehensive image-based table recognition. This integrated approach effectively handles diverse table styles, complex structures, and image distortions, resulting in improved accuracy and efficiency compared to existing methods like Table Transformers. Our system achieves simultaneous table detection (TD), table structure recognition (TSR), and table content recognition (TCR), preserving table structures and accurately extracting tabular data from document images. The integration of multiple models addresses the intricacies of table recognition, making our approach a promising solution for image-based table understanding, data extraction, and information retrieval applications. Our proposed approach achieves an IOU of 0.96 and an OCR Accuracy of 78%, showcasing a remarkable improvement of approximately 25% in the OCR Accuracy compared to the previous Table Transformer approach.

Avinash Anand, Raj Jaiswal, Mohit Gupta et al.

Large ground-truth datasets and recent advances in deep learning techniques have been useful for layout detection. However, because of the restricted layout diversity of these datasets, training on them requires a sizable number of annotated instances, which is both expensive and time-consuming. As a result, differences between the source and target domains may significantly impact how well these models function. To solve this problem, domain adaptation approaches have been developed that use a small quantity of labeled data to adjust the model to the target domain. In this research, we introduced a synthetic document dataset called RanLayNet, enriched with automatically assigned labels denoting spatial positions, ranges, and types of layout elements. The primary aim of this endeavor is to develop a versatile dataset capable of training models with robustness and adaptability to diverse document formats. Through empirical experimentation, we demonstrate that a deep layout identification model trained on our dataset exhibits enhanced performance compared to a model trained solely on actual documents. Moreover, we conduct a comparative analysis by fine-tuning inference models using both PubLayNet and IIIT-AR-13K datasets on the Doclaynet dataset. Our findings emphasize that models enriched with our dataset are optimal for tasks such as achieving 0.398 and 0.588 mAP95 score in the scientific document domain for the TABLE class.

Bhuiyan Sanjid Shafique, Ashmal Vayani, Muhammad Maaz et al.

Large multimodal models (LMMs) have recently gained attention due to their effectiveness to understand and generate descriptions of visual content. Most existing LMMs are in English language. While few recent works explore multilingual image LMMs, to the best of our knowledge, moving beyond the English language for cultural and linguistic inclusivity is yet to be investigated in the context of video LMMs. In pursuit of more inclusive video LMMs, we introduce a multilingual Video LMM benchmark, named ViMUL-Bench, to evaluate Video LMMs across 14 languages, including both low- and high-resource languages: English, Chinese, Spanish, French, German, Hindi, Arabic, Russian, Bengali, Urdu, Sinhala, Tamil, Swedish, and Japanese. Our ViMUL-Bench is designed to rigorously test video LMMs across 15 categories including eight culturally diverse categories, ranging from lifestyles and festivals to foods and rituals and from local landmarks to prominent cultural personalities. ViMUL-Bench comprises both open-ended (short and long-form) and multiple-choice questions spanning various video durations (short, medium, and long) with 8k samples that are manually verified by native language speakers. In addition, we also introduce a machine translated multilingual video training set comprising 1.2 million samples and develop a simple multilingual video LMM, named ViMUL, that is shown to provide a better tradeoff between high-and low-resource languages for video understanding. We hope our ViMUL-Bench and multilingual video LMM along with a large-scale multilingual video training set will help ease future research in developing cultural and linguistic inclusive multilingual video LMMs. Our proposed benchmark, video LMM and training data will be publicly released at https://mbzuai-oryx.github.io/ViMUL/.

Daniel Kienzle, Robin Schön, Rainer Lienhart et al.

Analyzing a player's technique in table tennis requires knowledge of the ball's 3D trajectory and spin. While, the spin is not directly observable in standard broadcasting videos, we show that it can be inferred from the ball's trajectory in the video. We present a novel method to infer the initial spin and 3D trajectory from the corresponding 2D trajectory in a video. Without ground truth labels for broadcast videos, we train a neural network solely on synthetic data. Due to the choice of our input data representation, physically correct synthetic training data, and using targeted augmentations, the network naturally generalizes to real data. Notably, these simple techniques are sufficient to achieve generalization. No real data at all is required for training. To the best of our knowledge, we are the first to present a method for spin and trajectory prediction in simple monocular broadcast videos, achieving an accuracy of 92.0% in spin classification and a 2D reprojection error of 0.19% of the image diagonal.

Ziling Huang, Yidan Zhang, Shin'ichi Satoh

Large-scale visual search engines are expected to solve a dual problem at once: (i) locate every image that truly contains the object described by a sentence and (ii) identify the object's bounding box or exact pixels within each hit. Existing techniques address only one side of this challenge. Visual grounding yields tight boxes and masks but rests on the unrealistic assumption that the object is present in every test image, producing a flood of false alarms when applied to web-scale collections. Text-to-image retrieval excels at sifting through massive databases to rank relevant images, yet it stops at whole-image matches and offers no fine-grained localization. We introduce Referring Search and Discovery (ReSeDis), the first task that unifies corpus-level retrieval with pixel-level grounding. Given a free-form description, a ReSeDis model must decide whether the queried object appears in each image and, if so, where it is, returning bounding boxes or segmentation masks. To enable rigorous study, we curate a benchmark in which every description maps uniquely to object instances scattered across a large, diverse corpus, eliminating unintended matches. We further design a task-specific metric that jointly scores retrieval recall and localization precision. Finally, we provide a straightforward zero-shot baseline using a frozen vision-language model, revealing significant headroom for future study. ReSeDis offers a realistic, end-to-end testbed for building the next generation of robust and scalable multimodal search systems.

Daniel Kienzle, Katja Ludwig, Julian Lorenz et al.

Obtaining the precise 3D motion of a table tennis ball from standard monocular videos is a challenging problem, as existing methods trained on synthetic data struggle to generalize to the noisy, imperfect ball and table detections of the real world. This is primarily due to the inherent lack of 3D ground truth trajectories and spin annotations for real-world video. To overcome this, we propose a novel two-stage pipeline that divides the problem into a front-end perception task and a back-end 2D-to-3D uplifting task. This separation allows us to train the front-end components with abundant 2D supervision from our newly created TTHQ dataset, while the back-end uplifting network is trained exclusively on physically-correct synthetic data. We specifically re-engineer the uplifting model to be robust to common real-world artifacts, such as missing detections and varying frame rates. By integrating a ball detector and a table keypoint detector, our approach transforms a proof-of-concept uplifting method into a practical, robust, and high-performing end-to-end application for 3D table tennis trajectory and spin analysis.

Haohua Dong, Ana Manzano Rodríguez, Camille Guinaudeau et al.

Face gender classification models often reflect and amplify demographic biases present in their training data, leading to uneven performance across gender and racial subgroups. We introduce pseudo-balancing, a simple and effective strategy for mitigating such biases in semi-supervised learning. Our method enforces demographic balance during pseudo-label selection, using only unlabeled images from a race-balanced dataset without requiring access to ground-truth annotations. We evaluate pseudo-balancing under two conditions: (1) fine-tuning a biased gender classifier using unlabeled images from the FairFace dataset, and (2) stress-testing the method with intentionally imbalanced training data to simulate controlled bias scenarios. In both cases, models are evaluated on the All-Age-Faces (AAF) benchmark, which contains a predominantly East Asian population. Our results show that pseudo-balancing consistently improves fairness while preserving or enhancing accuracy. The method achieves 79.81% overall accuracy - a 6.53% improvement over the baseline - and reduces the gender accuracy gap by 44.17%. In the East Asian subgroup, where baseline disparities exceeded 49%, the gap is narrowed to just 5.01%. These findings suggest that even in the absence of label supervision, access to a demographically balanced or moderately skewed unlabeled dataset can serve as a powerful resource for debiasing existing computer vision models.

Zhijing Wan, Zhixiang Wang, Zheng Wang et al.

One-shot subset selection serves as an effective tool to reduce deep learning training costs by identifying an informative data subset based on the information extracted by an information extractor (IE). Traditional IEs, typically pre-trained on the target dataset, are inherently dataset-dependent. Foundation models (FMs) offer a promising alternative, potentially mitigating this limitation. This work investigates two key questions: (1) Can FM-based subset selection outperform traditional IE-based methods across diverse datasets? (2) Do all FMs perform equally well as IEs for subset selection? Extensive experiments uncovered surprising insights: FMs consistently outperform traditional IEs on fine-grained datasets, whereas their advantage diminishes on coarse-grained datasets with noisy labels. Motivated by these finding, we propose RAM-APL (RAnking Mean-Accuracy of Pseudo-class Labels), a method tailored for fine-grained image datasets. RAM-APL leverages multiple FMs to enhance subset selection by exploiting their complementary strengths. Our approach achieves state-of-the-art performance on fine-grained datasets, including Oxford-IIIT Pet, Food-101, and Caltech-UCSD Birds-200-2011.

Rintaro Yanagi, Yamato Okamoto, Shuhei Yokoo et al.

The short video-sharing services that allow users to post 10-30 second videos (e.g., YouTube Shorts and TikTok) have attracted a lot of attention in recent years. However, conventional video copy detection (VCD) methods mainly focus on general video-sharing services (e.g., YouTube and Bilibili), and the effects of short video-sharing services on video copy detection are still unclear. Considering that illegally copied videos in short video-sharing services have service-distinctive characteristics, especially in those time lengths, the pros and cons of VCD in those services are required to be analyzed. In this paper, we examine the effects of short video-sharing services on VCD by constructing a dataset that has short video-sharing service characteristics. Our novel dataset is automatically constructed from the publicly available dataset to have reference videos and fixed short-time-length query videos, and such automation procedures assure the reproducibility and data privacy preservation of this paper. From the experimental results focusing on segment-level and video-level situations, we can see that three effects: "Segment-level VCD in short video-sharing services is more difficult than those in general video-sharing services", "Video-level VCD in short video-sharing services is easier than those in general video-sharing services", "The video alignment component mainly suppress the detection performance in short video-sharing services".

Nobukatsu Kajiura, Hong Liu, Shin'ichi Satoh

This paper focuses on camouflaged object detection (COD), which is a task to detect objects hidden in the background. Most of the current COD models aim to highlight the target object directly while outputting ambiguous camouflaged boundaries. On the other hand, the performance of the models considering edge information is not yet satisfactory. To this end, we propose a new framework that makes full use of multiple visual cues, i.e., saliency as well as edges, to refine the predicted camouflaged map. This framework consists of three key components, i.e., a pseudo-edge generator, a pseudo-map generator, and an uncertainty-aware refinement module. In particular, the pseudo-edge generator estimates the boundary that outputs the pseudo-edge label, and the conventional COD method serves as the pseudo-map generator that outputs the pseudo-map label. Then, we propose an uncertainty-based module to reduce the uncertainty and noise of such two pseudo labels, which takes both pseudo labels as input and outputs an edge-accurate camouflaged map. Experiments on various COD datasets demonstrate the effectiveness of our method with superior performance to the existing state-of-the-art methods.

Riku Togashi, Masahiro Kato, Mayu Otani et al.

Learning from implicit feedback is challenging because of the difficult nature of the one-class problem: we can observe only positive examples. Most conventional methods use a pairwise ranking approach and negative samplers to cope with the one-class problem. However, such methods have two main drawbacks particularly in large-scale applications; (1) the pairwise approach is severely inefficient due to the quadratic computational cost; and (2) even recent model-based samplers (e.g. IRGAN) cannot achieve practical efficiency due to the training of an extra model. In this paper, we propose a learning-to-rank approach, which achieves convergence speed comparable to the pointwise counterpart while performing similarly to the pairwise counterpart in terms of ranking effectiveness. Our approach estimates the probability densities of positive items for each user within a rich class of distributions, viz. \emph{exponential family}. In our formulation, we derive a loss function and the appropriate negative sampling distribution based on maximum likelihood estimation. We also develop a practical technique for risk approximation and a regularisation scheme. We then discuss that our single-model approach is equivalent to an IRGAN variant under a certain condition. Through experiments on real-world datasets, our approach outperforms the pointwise and pairwise counterparts in terms of effectiveness and efficiency.

Riku Togashi, Masahiro Kato, Mayu Otani et al.

Learning from implicit user feedback is challenging as we can only observe positive samples but never access negative ones. Most conventional methods cope with this issue by adopting a pairwise ranking approach with negative sampling. However, the pairwise ranking approach has a severe disadvantage in the convergence time owing to the quadratically increasing computational cost with respect to the sample size; it is problematic, particularly for large-scale datasets and complex models such as neural networks. By contrast, a pointwise approach does not directly solve a ranking problem, and is therefore inferior to a pairwise counterpart in top-K ranking tasks; however, it is generally advantageous in regards to the convergence time. This study aims to establish an approach to learn personalised ranking from implicit feedback, which reconciles the training efficiency of the pointwise approach and ranking effectiveness of the pairwise counterpart. The key idea is to estimate the ranking of items in a pointwise manner; we first reformulate the conventional pointwise approach based on density ratio estimation and then incorporate the essence of ranking-oriented approaches (e.g. the pairwise approach) into our formulation. Through experiments on three real-world datasets, we demonstrate that our approach not only dramatically reduces the convergence time (one to two orders of magnitude faster) but also significantly improving the ranking performance.

Liang Liao, Jing Xiao, Zheng Wang et al.

Existing inpainting methods have achieved promising performance in recovering defected images of specific scenes. However, filling holes involving multiple semantic categories remains challenging due to the obscure semantic boundaries and the mixture of different semantic textures. In this paper, we introduce coherence priors between the semantics and textures which make it possible to concentrate on completing separate textures in a semantic-wise manner. Specifically, we adopt a multi-scale joint optimization framework to first model the coherence priors and then accordingly interleavingly optimize image inpainting and semantic segmentation in a coarse-to-fine manner. A Semantic-Wise Attention Propagation (SWAP) module is devised to refine completed image textures across scales by exploring non-local semantic coherence, which effectively mitigates mix-up of textures. We also propose two coherence losses to constrain the consistency between the semantics and the inpainted image in terms of the overall structure and detailed textures. Experimental results demonstrate the superiority of our proposed method for challenging cases with complex holes.

Riku Togashi, Mayu Otani, Shin'ichi Satoh

Solving cold-start problems is indispensable to provide meaningful recommendation results for new users and items. Under sparsely observed data, unobserved user-item pairs are also a vital source for distilling latent users' information needs. Most present works leverage unobserved samples for extracting negative signals. However, such an optimisation strategy can lead to biased results toward already popular items by frequently handling new items as negative instances. In this study, we tackle the cold-start problems for new users/items by appropriately leveraging unobserved samples. We propose a knowledge graph (KG)-aware recommender based on graph neural networks, which augments labelled samples through pseudo-labelling. Our approach aggressively employs unobserved samples as positive instances and brings new items into the spotlight. To avoid exhaustive label assignments to all possible pairs of users and items, we exploit a KG for selecting probably positive items for each user. We also utilise an improved negative sampling strategy and thereby suppress the exacerbation of popularity biases. Through experiments, we demonstrate that our approach achieves improvements over the state-of-the-art KG-aware recommenders in a variety of scenarios; in particular, our methodology successfully improves recommendation performance for cold-start users/items.

Yuting Liu, Zheng Wang, Miaojing Shi et al.

Unsupervised crowd counting is a challenging yet not largely explored task. In this paper, we explore it in a transfer learning setting where we learn to detect and count persons in an unlabeled target set by transferring bi-knowledge learnt from regression- and detection-based models in a labeled source set. The dual source knowledge of the two models is heterogeneous and complementary as they capture different modalities of the crowd distribution. We formulate the mutual transformations between the outputs of regression- and detection-based models as two scene-agnostic transformers which enable knowledge distillation between the two models. Given the regression- and detection-based models and their mutual transformers learnt in the source, we introduce an iterative self-supervised learning scheme with regression-detection bi-knowledge transfer in the target. Extensive experiments on standard crowd counting benchmarks, ShanghaiTech, UCF\_CC\_50, and UCF\_QNRF demonstrate a substantial improvement of our method over other state-of-the-arts in the transfer learning setting.

Liang Liao, Jing Xiao, Zheng Wang et al.

Completing a corrupted image with correct structures and reasonable textures for a mixed scene remains an elusive challenge. Since the missing hole in a mixed scene of a corrupted image often contains various semantic information, conventional two-stage approaches utilizing structural information often lead to the problem of unreliable structural prediction and ambiguous image texture generation. In this paper, we propose a Semantic Guidance and Evaluation Network (SGE-Net) to iteratively update the structural priors and the inpainted image in an interplay framework of semantics extraction and image inpainting. It utilizes semantic segmentation map as guidance in each scale of inpainting, under which location-dependent inferences are re-evaluated, and, accordingly, poorly-inferred regions are refined in subsequent scales. Extensive experiments on real-world images of mixed scenes demonstrated the superiority of our proposed method over state-of-the-art approaches, in terms of clear boundaries and photo-realistic textures.

Changhee Han, Leonardo Rundo, Kohei Murao et al.

Unsupervised learning can discover various unseen diseases, relying on large-scale unannotated medical images of healthy subjects. Towards this, unsupervised methods reconstruct a single medical image to detect outliers either in the learned feature space or from high reconstruction loss. However, without considering continuity between multiple adjacent slices, they cannot directly discriminate diseases composed of the accumulation of subtle anatomical anomalies, such as Alzheimer's Disease (AD). Moreover, no study has shown how unsupervised anomaly detection is associated with disease stages. Therefore, we propose a two-step method using Generative Adversarial Network-based multiple adjacent brain MRI slice reconstruction to detect AD at various stages: (Reconstruction) Wasserstein loss with Gradient Penalty + L1 loss---trained on 3 healthy slices to reconstruct the next 3 ones---reconstructs unseen healthy/AD cases; (Diagnosis) Average/Maximum loss (e.g., L2 loss) per scan discriminates them, comparing the reconstructed/ground truth images. The results show that we can reliably detect AD at a very early stage with Area Under the Curve (AUC) 0.780 while also detecting AD at a late stage much more accurately with AUC 0.917; since our method is fully unsupervised, it should also discover and alert any anomalies including rare disease.

Ziling Huang, Zheng Wang, Chung-Chi Tsai et al.

Group re-identification (G-ReID) is an important yet less-studied task. Its challenges not only lie in appearance changes of individuals which have been well-investigated in general person re-identification (ReID), but also derive from group layout and membership changes. So the key task of G-ReID is to learn representations robust to such changes. To address this issue, we propose a Transferred Single and Couple Representation Learning Network (TSCN). Its merits are two aspects: 1) Due to the lack of labelled training samples, existing G-ReID methods mainly rely on unsatisfactory hand-crafted features. To gain the superiority of deep learning models, we treat a group as multiple persons and transfer the domain of a labeled ReID dataset to a G-ReID target dataset style to learn single representations. 2) Taking into account the neighborhood relationship in a group, we further propose learning a novel couple representation between two group members, that achieves more discriminative power in G-ReID tasks. In addition, an unsupervised weight learning method is exploited to adaptively fuse the results of different views together according to result patterns. Extensive experimental results demonstrate the effectiveness of our approach that significantly outperforms state-of-the-art methods by 11.7\% CMC-1 on the Road Group dataset and by 39.0\% CMC-1 on the DukeMCMT dataset.

Zelong Zeng, Zhixiang Wang, Zheng Wang et al.

Most person re-identification (ReID) approaches assume that person images are captured under relatively similar illumination conditions. In reality, long-term person retrieval is common, and person images are often captured under different illumination conditions at different times across a day. In this situation, the performances of existing ReID models often degrade dramatically. This paper addresses the ReID problem with illumination variations and names it as {\em Illumination-Adaptive Person Re-identification (IA-ReID)}. We propose an Illumination-Identity Disentanglement (IID) network to dispel different scales of illuminations away while preserving individuals' identity information. To demonstrate the illumination issue and to evaluate our model, we construct two large-scale simulated datasets with a wide range of illumination variations. Experimental results on the simulated datasets and real-world images demonstrate the effectiveness of the proposed framework.

Changhee Han, Kohei Murao, Shin'ichi Satoh et al.

Convolutional Neural Network (CNN)-based accurate prediction typically requires large-scale annotated training data. In Medical Imaging, however, both obtaining medical data and annotating them by expert physicians are challenging; to overcome this lack of data, Data Augmentation (DA) using Generative Adversarial Networks (GANs) is essential, since they can synthesize additional annotated training data to handle small and fragmented medical images from various scanners--those generated images, realistic but completely novel, can further fill the real image distribution uncovered by the original dataset. As a tutorial, this paper introduces GAN-based Medical Image Augmentation, along with tricks to boost classification/object detection/segmentation performance using them, based on our experience and related work. Moreover, we show our first GAN-based DA work using automatic bounding box annotation, for robust CNN-based brain metastases detection on 256 x 256 MR images; GAN-based DA can boost 10% sensitivity in diagnosis with a clinically acceptable number of additional False Positives, even with highly-rough and inconsistent bounding boxes.

Changhee Han, Kohei Murao, Tomoyuki Noguchi et al.

Accurate Computer-Assisted Diagnosis, associated with proper data wrangling, can alleviate the risk of overlooking the diagnosis in a clinical environment. Towards this, as a Data Augmentation (DA) technique, Generative Adversarial Networks (GANs) can synthesize additional training data to handle the small/fragmented medical imaging datasets collected from various scanners; those images are realistic but completely different from the original ones, filling the data lack in the real image distribution. However, we cannot easily use them to locate disease areas, considering expert physicians' expensive annotation cost. Therefore, this paper proposes Conditional Progressive Growing of GANs (CPGGANs), incorporating highly-rough bounding box conditions incrementally into PGGANs to place brain metastases at desired positions/sizes on 256 X 256 Magnetic Resonance (MR) images, for Convolutional Neural Network-based tumor detection; this first GAN-based medical DA using automatic bounding box annotation improves the training robustness. The results show that CPGGAN-based DA can boost 10% sensitivity in diagnosis with clinically acceptable additional False Positives. Surprisingly, further tumor realism, achieved with additional normal brain MR images for CPGGAN training, does not contribute to detection performance, while even three physicians cannot accurately distinguish them from the real ones in Visual Turing Test.

Fan Yang, Ryota Hinami, Yusuke Matsui et al.

Diffusion is commonly used as a ranking or re-ranking method in retrieval tasks to achieve higher retrieval performance, and has attracted lots of attention in recent years. A downside to diffusion is that it performs slowly in comparison to the naive k-NN search, which causes a non-trivial online computational cost on large datasets. To overcome this weakness, we propose a novel diffusion technique in this paper. In our work, instead of applying diffusion to the query, we pre-compute the diffusion results of each element in the database, making the online search a simple linear combination on top of the k-NN search process. Our proposed method becomes 10~ times faster in terms of online search speed. Moreover, we propose to use late truncation instead of early truncation in previous works to achieve better retrieval performance.

Yusuke Matsui, Ryota Hinami, Shin'ichi Satoh

Existing approximate nearest neighbor search systems suffer from two fundamental problems that are of practical importance but have not received sufficient attention from the research community. First, although existing systems perform well for the whole database, it is difficult to run a search over a subset of the database. Second, there has been no discussion concerning the performance decrement after many items have been newly added to a system. We develop a reconfigurable inverted index (Rii) to resolve these two issues. Based on the standard IVFADC system, we design a data layout such that items are stored linearly. This enables us to efficiently run a subset search by switching the search method to a linear PQ scan if the size of a subset is small. Owing to the linear layout, the data structure can be dynamically adjusted after new items are added, maintaining the fast speed of the system. Extensive comparisons show that Rii achieves a comparable performance with state-of-the art systems such as Faiss.

Yaman Kumar, Mayank Aggarwal, Pratham Nawal et al.

Speechreading or lipreading is the technique of understanding and getting phonetic features from a speaker's visual features such as movement of lips, face, teeth and tongue. It has a wide range of multimedia applications such as in surveillance, Internet telephony, and as an aid to a person with hearing impairments. However, most of the work in speechreading has been limited to text generation from silent videos. Recently, research has started venturing into generating (audio) speech from silent video sequences but there have been no developments thus far in dealing with divergent views and poses of a speaker. Thus although, we have multiple camera feeds for the speech of a user, but we have failed in using these multiple video feeds for dealing with the different poses. To this end, this paper presents the world's first ever multi-view speech reading and reconstruction system. This work encompasses the boundaries of multimedia research by putting forth a model which leverages silent video feeds from multiple cameras recording the same subject to generate intelligent speech for a speaker. Initial results confirm the usefulness of exploiting multiple camera views in building an efficient speech reading and reconstruction system. It further shows the optimal placement of cameras which would lead to the maximum intelligibility of speech. Next, it lays out various innovative applications for the proposed system focusing on its potential prodigious impact in not just security arena but in many other multimedia analytics problems.

Ryota Hinami, Junwei Liang, Shin'ichi Satoh et al.

We tackle the problem of learning concept classifiers from videos on the web without using manually labeled data. Although metadata attached to videos (e.g., video titles, descriptions) can be of help collecting training data for the target concept, the collected data is often very noisy. The main challenge is therefore how to select good examples from noisy training data. Previous approaches firstly learn easy examples that are unlikely to be noise and then gradually learn more complex examples. However, hard examples that are much different from easy ones are never learned. In this paper, we propose an approach called multimodal co-training (MMCo) for selecting good examples from noisy training data. MMCo jointly learns classifiers for multiple modalities that complement each other to select good examples. Since MMCo selects examples by consensus of multimodal classifiers, a hard example for one modality can still be used as a training example by exploiting the power of the other modalities. The algorithm is very simple and easily implemented but yields consistent and significant boosts in example selection and classification performance on the FCVID and YouTube8M benchmarks.

Andreu Girbau, Ryota Hinami, Shin'ichi Satoh

In this work we propose tracking as a generic addition to the instance search task. From video data perspective, much information that can be used is not taken into account in the traditional instance search approach. This work aims to provide insights on exploiting such existing information by means of tracking and the proper combination of the results, independently of the instance search system. We also present a study on the improvement of the system when using multiple independent instances (up to 4) of the same person. Experimental results show that our system improves substantially its performance when using tracking. Best configuration improves from mAP = 0.447 to mAP = 0.511 for a single example, and from mAP = 0.647 to mAP = 0.704 for multiple (4) given examples.

Yuki Nagai, Yusuke Uchida, Shigeyuki Sakazawa et al.

Although deep neural networks have made tremendous progress in the area of multimedia representation, training neural models requires a large amount of data and time. It is well-known that utilizing trained models as initial weights often achieves lower training error than neural networks that are not pre-trained. A fine-tuning step helps to reduce both the computational cost and improve performance. Therefore, sharing trained models has been very important for the rapid progress of research and development. In addition, trained models could be important assets for the owner(s) who trained them, hence we regard trained models as intellectual property. In this paper, we propose a digital watermarking technology for ownership authorization of deep neural networks. First, we formulate a new problem: embedding watermarks into deep neural networks. We also define requirements, embedding situations, and attack types on watermarking in deep neural networks. Second, we propose a general framework for embedding a watermark in model parameters, using a parameter regularizer. Our approach does not impair the performance of networks into which a watermark is placed because the watermark is embedded while training the host network. Finally, we perform comprehensive experiments to reveal the potential of watermarking deep neural networks as the basis of this new research effort. We show that our framework can embed a watermark during the training of a deep neural network from scratch, and during fine-tuning and distilling, without impairing its performance. The embedded watermark does not disappear even after fine-tuning or parameter pruning; the watermark remains complete even after 65% of parameters are pruned.

Sang Phan, Gustav Eje Henter, Yusuke Miyao et al.

Captioning models are typically trained using the cross-entropy loss. However, their performance is evaluated on other metrics designed to better correlate with human assessments. Recently, it has been shown that reinforcement learning (RL) can directly optimize these metrics in tasks such as captioning. However, this is computationally costly and requires specifying a baseline reward at each step to make training converge. We propose a fast approach to optimize one's objective of interest through the REINFORCE algorithm. First we show that, by replacing model samples with ground-truth sentences, RL training can be seen as a form of weighted cross-entropy loss, giving a fast, RL-based pre-training algorithm. Second, we propose to use the consensus among ground-truth captions of the same video as the baseline reward. This can be computed very efficiently. We call the complete proposal Consensus-based Sequence Training (CST). Applied to the MSRVTT video captioning benchmark, our proposals train significantly faster than comparable methods and establish a new state-of-the-art on the task, improving the CIDEr score from 47.3 to 54.2.

Ryota Hinami, Shin'ichi Satoh

Thanks to the success of object detection technology, we can retrieve objects of the specified classes even from huge image collections. However, the current state-of-the-art object detectors (such as Faster R-CNN) can only handle pre-specified classes. In addition, large amounts of positive and negative visual samples are required for training. In this paper, we address the problem of open-vocabulary object retrieval and localization, where the target object is specified by a textual query (e.g., a word or phrase). We first propose Query-Adaptive R-CNN, a simple extension of Faster R-CNN adapted to open-vocabulary queries, by transforming the text embedding vector into an object classifier and localization regressor. Then, for discriminative training, we then propose negative phrase augmentation (NPA) to mine hard negative samples which are visually similar to the query and at the same time semantically mutually exclusive of the query. The proposed method can retrieve and localize objects specified by a textual query from one million images in only 0.5 seconds with high precision.

Ryota Hinami, Tao Mei, Shin'ichi Satoh

This paper addresses the problem of joint detection and recounting of abnormal events in videos. Recounting of abnormal events, i.e., explaining why they are judged to be abnormal, is an unexplored but critical task in video surveillance, because it helps human observers quickly judge if they are false alarms or not. To describe the events in the human-understandable form for event recounting, learning generic knowledge about visual concepts (e.g., object and action) is crucial. Although convolutional neural networks (CNNs) have achieved promising results in learning such concepts, it remains an open question as to how to effectively use CNNs for abnormal event detection, mainly due to the environment-dependent nature of the anomaly detection. In this paper, we tackle this problem by integrating a generic CNN model and environment-dependent anomaly detectors. Our approach first learns CNN with multiple visual tasks to exploit semantic information that is useful for detecting and recounting abnormal events. By appropriately plugging the model into anomaly detectors, we can detect and recount abnormal events while taking advantage of the discriminative power of CNNs. Our approach outperforms the state-of-the-art on Avenue and UCSD Ped2 benchmarks for abnormal event detection and also produces promising results of abnormal event recounting.