Dong-Jin Kim

Taehoon Kim, Pyunghwan Ahn, Sangyun Kim et al. · nvidia, utoronto

In this report, we introduce NICE (New frontiers for zero-shot Image Captioning Evaluation) project and share the results and outcomes of 2023 challenge. This project is designed to challenge the computer vision community to develop robust image captioning models that advance the state-of-the-art both in terms of accuracy and fairness. Through the challenge, the image captioning models were tested using a new evaluation dataset that includes a large variety of visual concepts from many domains. There was no specific training data provided for the challenge, and therefore the challenge entries were required to adapt to new types of image descriptions that had not been seen during training. This report includes information on the newly proposed NICE dataset, evaluation methods, challenge results, and technical details of top-ranking entries. We expect that the outcomes of the challenge will contribute to the improvement of AI models on various vision-language tasks.

Jae Won Cho, Dong-jin Kim, Hyeonggon Ryu et al.

The task of Visual Question Answering (VQA) is known to be plagued by the issue of VQA models exploiting biases within the dataset to make its final prediction. Various previous ensemble based debiasing methods have been proposed where an additional model is purposefully trained to be biased in order to train a robust target model. However, these methods compute the bias for a model simply from the label statistics of the training data or from single modal branches. In this work, in order to better learn the bias a target VQA model suffers from, we propose a generative method to train the bias model directly from the target model, called GenB. In particular, GenB employs a generative network to learn the bias in the target model through a combination of the adversarial objective and knowledge distillation. We then debias our target model with GenB as a bias model, and show through extensive experiments the effects of our method on various VQA bias datasets including VQA-CP2, VQA-CP1, GQA-OOD, and VQA-CE, and show state-of-the-art results with the LXMERT architecture on VQA-CP2.

Youngjoon Jang, Youngtaek Oh, Jae Won Cho et al.

The goal of this work is background-robust continuous sign language recognition. Most existing Continuous Sign Language Recognition (CSLR) benchmarks have fixed backgrounds and are filmed in studios with a static monochromatic background. However, signing is not limited only to studios in the real world. In order to analyze the robustness of CSLR models under background shifts, we first evaluate existing state-of-the-art CSLR models on diverse backgrounds. To synthesize the sign videos with a variety of backgrounds, we propose a pipeline to automatically generate a benchmark dataset utilizing existing CSLR benchmarks. Our newly constructed benchmark dataset consists of diverse scenes to simulate a real-world environment. We observe even the most recent CSLR method cannot recognize glosses well on our new dataset with changed backgrounds. In this regard, we also propose a simple yet effective training scheme including (1) background randomization and (2) feature disentanglement for CSLR models. The experimental results on our dataset demonstrate that our method generalizes well to other unseen background data with minimal additional training images.

Youngjoon Jang, Youngtaek Oh, Jae Won Cho et al.

The goal of this work is to develop self-sufficient framework for Continuous Sign Language Recognition (CSLR) that addresses key issues of sign language recognition. These include the need for complex multi-scale features such as hands, face, and mouth for understanding, and absence of frame-level annotations. To this end, we propose (1) Divide and Focus Convolution (DFConv) which extracts both manual and non-manual features without the need for additional networks or annotations, and (2) Dense Pseudo-Label Refinement (DPLR) which propagates non-spiky frame-level pseudo-labels by combining the ground truth gloss sequence labels with the predicted sequence. We demonstrate that our model achieves state-of-the-art performance among RGB-based methods on large-scale CSLR benchmarks, PHOENIX-2014 and PHOENIX-2014-T, while showing comparable results with better efficiency when compared to other approaches that use multi-modality or extra annotations.

Muhammad Kashif Ali, Dongjin Kim, Tae Hyun Kim

In many video restoration/translation tasks, image processing operations are naïvely extended to the video domain by processing each frame independently, disregarding the temporal connection of the video frames. This disregard for the temporal connection often leads to severe temporal inconsistencies. State-Of-The-Art (SOTA) techniques that address these inconsistencies rely on the availability of unprocessed videos to implicitly siphon and utilize consistent video dynamics to restore the temporal consistency of frame-wise processed videos which often jeopardizes the translation effect. We propose a general framework for this task that learns to infer and utilize consistent motion dynamics from inconsistent videos to mitigate the temporal flicker while preserving the perceptual quality for both the temporally neighboring and relatively distant frames without requiring the raw videos at test time. The proposed framework produces SOTA results on two benchmark datasets, DAVIS and videvo.net, processed by numerous image processing applications. The code and the trained models are available at \url{https://github.com/MKashifAli/TARONVD}.

Dong-Jin Kim, Tae-Hyun Oh, Jinsoo Choi et al.

We present a novel data-efficient semi-supervised framework to improve the generalization of image captioning models. Constructing a large-scale labeled image captioning dataset is an expensive task in terms of labor, time, and cost. In contrast to manually annotating all the training samples, separately collecting uni-modal datasets is immensely easier, e.g., a large-scale image dataset and a sentence dataset. We leverage such massive unpaired image and caption data upon standard paired data by learning to associate them. To this end, our proposed semi-supervised learning method assigns pseudo-labels to unpaired samples in an adversarial learning fashion, where the joint distribution of image and caption is learned. Our method trains a captioner to learn from a paired data and to progressively associate unpaired data. This approach shows noticeable performance improvement even in challenging scenarios including out-of-task data (i.e., relational captioning, where the target task is different from the unpaired data) and web-crawled data. We also show that our proposed method is theoretically well-motivated and has a favorable global optimal property. Our extensive and comprehensive empirical results both on (1) image-based and (2) dense region-based captioning datasets followed by comprehensive analysis on the scarcely-paired COCO dataset demonstrate the consistent effectiveness of our semisupervised learning method with unpaired data compared to competing methods.

Sung Jin Um, Dongjin Kim, Jung Uk Kim

The objective of the sound source localization task is to enable machines to detect the location of sound-making objects within a visual scene. While the audio modality provides spatial cues to locate the sound source, existing approaches only use audio as an auxiliary role to compare spatial regions of the visual modality. Humans, on the other hand, utilize both audio and visual modalities as spatial cues to locate sound sources. In this paper, we propose an audio-visual spatial integration network that integrates spatial cues from both modalities to mimic human behavior when detecting sound-making objects. Additionally, we introduce a recursive attention network to mimic human behavior of iterative focusing on objects, resulting in more accurate attention regions. To effectively encode spatial information from both modalities, we propose audio-visual pair matching loss and spatial region alignment loss. By utilizing the spatial cues of audio-visual modalities and recursively focusing objects, our method can perform more robust sound source localization. Comprehensive experimental results on the Flickr SoundNet and VGG-Sound Source datasets demonstrate the superiority of our proposed method over existing approaches. Our code is available at: https://github.com/VisualAIKHU/SIRA-SSL

Dongjin Kim, Sung Jin Um, Sangmin Lee et al.

The goal of the multi-sound source localization task is to localize sound sources from the mixture individually. While recent multi-sound source localization methods have shown improved performance, they face challenges due to their reliance on prior information about the number of objects to be separated. In this paper, to overcome this limitation, we present a novel multi-sound source localization method that can perform localization without prior knowledge of the number of sound sources. To achieve this goal, we propose an iterative object identification (IOI) module, which can recognize sound-making objects in an iterative manner. After finding the regions of sound-making objects, we devise object similarity-aware clustering (OSC) loss to guide the IOI module to effectively combine regions of the same object but also distinguish between different objects and backgrounds. It enables our method to perform accurate localization of sound-making objects without any prior knowledge. Extensive experimental results on the MUSIC and VGGSound benchmarks show the significant performance improvements of the proposed method over the existing methods for both single and multi-source. Our code is available at: https://github.com/VisualAIKHU/NoPrior_MultiSSL

Soeun Lee, Si-Woo Kim, Taewhan Kim et al.

Recent advancements in image captioning have explored text-only training methods to overcome the limitations of paired image-text data. However, existing text-only training methods often overlook the modality gap between using text data during training and employing images during inference. To address this issue, we propose a novel approach called Image-like Retrieval, which aligns text features with visually relevant features to mitigate the modality gap. Our method further enhances the accuracy of generated captions by designing a Fusion Module that integrates retrieved captions with input features. Additionally, we introduce a Frequency-based Entity Filtering technique that significantly improves caption quality. We integrate these methods into a unified framework, which we refer to as IFCap ($\textbf{I}$mage-like Retrieval and $\textbf{F}$requency-based Entity Filtering for Zero-shot $\textbf{Cap}$tioning). Through extensive experimentation, our straightforward yet powerful approach has demonstrated its efficacy, outperforming the state-of-the-art methods by a significant margin in both image captioning and video captioning compared to zero-shot captioning based on text-only training.

Taewhan Kim, Soeun Lee, Si-Woo Kim et al.

Recent lightweight image captioning models using retrieved data mainly focus on text prompts. However, previous works only utilize the retrieved text as text prompts, and the visual information relies only on the CLIP visual embedding. Because of this issue, there is a limitation that the image descriptions inherent in the prompt are not sufficiently reflected in the visual embedding space. To tackle this issue, we propose ViPCap, a novel retrieval text-based visual prompt for lightweight image captioning. ViPCap leverages the retrieved text with image information as visual prompts to enhance the ability of the model to capture relevant visual information. By mapping text prompts into the CLIP space and generating multiple randomized Gaussian distributions, our method leverages sampling to explore randomly augmented distributions and effectively retrieves the semantic features that contain image information. These retrieved features are integrated into the image and designated as the visual prompt, leading to performance improvements on the datasets such as COCO, Flickr30k, and NoCaps. Experimental results demonstrate that ViPCap significantly outperforms prior lightweight captioning models in efficiency and effectiveness, demonstrating the potential for a plug-and-play solution. The source code is available at https://github.com/taewhankim/VIPCAP.

Muhammad Kashif Ali, Eun Woo Im, Dongjin Kim et al.

Video stabilization is a longstanding computer vision problem, particularly pixel-level synthesis solutions for video stabilization which synthesize full frames add to the complexity of this task. These techniques aim to stabilize videos by synthesizing full frames while enhancing the stability of the considered video. This intensifies the complexity of the task due to the distinct mix of unique motion profiles and visual content present in each video sequence, making robust generalization with fixed parameters difficult. In our study, we introduce a novel approach to enhance the performance of pixel-level synthesis solutions for video stabilization by adapting these models to individual input video sequences. The proposed adaptation exploits low-level visual cues accessible during test-time to improve both the stability and quality of resulting videos. We highlight the efficacy of our methodology of "test-time adaptation" through simple fine-tuning of one of these models, followed by significant stability gain via the integration of meta-learning techniques. Notably, significant improvement is achieved with only a single adaptation step. The versatility of the proposed algorithm is demonstrated by consistently improving the performance of various pixel-level synthesis models for video stabilization in real-world scenarios.

Si-Woo Kim, MinJu Jeon, Ye-Chan Kim et al.

Zero-shot Image Captioning (ZIC) increasingly utilizes synthetic datasets generated by text-to-image (T2I) models to mitigate the need for costly manual annotation. However, these T2I models often produce images that exhibit semantic misalignments with their corresponding input captions (e.g., missing objects, incorrect attributes), resulting in noisy synthetic image-caption pairs that can hinder model training. Existing dataset pruning techniques are largely designed for removing noisy text in web-crawled data. However, these methods are ill-suited for the distinct challenges of synthetic data, where captions are typically well-formed, but images may be inaccurate representations. To address this gap, we introduce SynC, a novel framework specifically designed to refine synthetic image-caption datasets for ZIC. Instead of conventional filtering or regeneration, SynC focuses on reassigning captions to the most semantically aligned images already present within the synthetic image pool. Our approach employs a one-to-many mapping strategy by initially retrieving multiple relevant candidate images for each caption. We then apply a cycle-consistency-inspired alignment scorer that selects the best image by verifying its ability to retrieve the original caption via image-to-text retrieval. Extensive evaluations demonstrate that SynC consistently and significantly improves performance across various ZIC models on standard benchmarks (MS-COCO, Flickr30k, NoCaps), achieving state-of-the-art results in several scenarios. SynC offers an effective strategy for curating refined synthetic data to enhance ZIC.

Mincheol Park, Dongjin Kim, Cheonjun Park et al.

Channel pruning is widely accepted to accelerate modern convolutional neural networks (CNNs). The resulting pruned model benefits from its immediate deployment on general-purpose software and hardware resources. However, its large pruning granularity, specifically at the unit of a convolution filter, often leads to undesirable accuracy drops due to the inflexibility of deciding how and where to introduce sparsity to the CNNs. In this paper, we propose REPrune, a novel channel pruning technique that emulates kernel pruning, fully exploiting the finer but structured granularity. REPrune identifies similar kernels within each channel using agglomerative clustering. Then, it selects filters that maximize the incorporation of kernel representatives while optimizing the maximum cluster coverage problem. By integrating with a simultaneous training-pruning paradigm, REPrune promotes efficient, progressive pruning throughout training CNNs, avoiding the conventional train-prune-finetune sequence. Experimental results highlight that REPrune performs better in computer vision tasks than existing methods, effectively achieving a balance between acceleration ratio and performance retention.

MinJu Jeon, Si-Woo Kim, Ye-Chan Kim et al.

Dense video captioning aims to temporally localize events in video and generate captions for each event. While recent works propose end-to-end models, they suffer from two limitations: (1) applying timestamp supervision only to text while treating all video frames equally, and (2) retrieving captions from fixed-size video chunks, overlooking scene transitions. To address these, we propose Sali4Vid, a simple yet effective saliency-aware framework. We introduce Saliency-aware Video Reweighting, which converts timestamp annotations into sigmoid-based frame importance weights, and Semantic-based Adaptive Caption Retrieval, which segments videos by frame similarity to capture scene transitions and improve caption retrieval. Sali4Vid achieves state-of-the-art results on YouCook2 and ViTT, demonstrating the benefit of jointly improving video weighting and retrieval for dense video captioning

Min-Jung Kim, Dongjin Kim, Seokju Yun et al.

Video editing has garnered increasing attention alongside the rapid progress of diffusion-based video generation models. As part of these advancements, there is a growing demand for more accessible and controllable forms of video editing, such as prompt-based editing. Previous studies have primarily focused on tasks such as style transfer, background replacement, object substitution, and attribute modification, while maintaining the content structure of the source video. However, more complex tasks, including the addition of novel objects and nonrigid transformations, remain relatively unexplored. In this paper, we present TV-LiVE, a Training-free and text-guided Video editing framework via Layerinformed Vitality Exploitation. We empirically identify vital layers within the video generation model that significantly influence the quality of generated outputs. Notably, these layers are closely associated with Rotary Position Embeddings (RoPE). Based on this observation, our method enables both object addition and non-rigid video editing by selectively injecting key and value features from the source model into the corresponding layers of the target model guided by the layer vitality. For object addition, we further identify prominent layers to extract the mask regions corresponding to the newly added target prompt. We found that the extracted masks from the prominent layers faithfully indicate the region to be edited. Experimental results demonstrate that TV-LiVE outperforms existing approaches for both object addition and non-rigid video editing. Project Page: https://emjay73.github.io/TV_LiVE/

Dong-Jin Kim, Jae Won Cho, Jinsoo Choi et al.

Constructing a large-scale labeled dataset in the real world, especially for high-level tasks (eg, Visual Question Answering), can be expensive and time-consuming. In addition, with the ever-growing amounts of data and architecture complexity, Active Learning has become an important aspect of computer vision research. In this work, we address Active Learning in the multi-modal setting of Visual Question Answering (VQA). In light of the multi-modal inputs, image and question, we propose a novel method for effective sample acquisition through the use of ad hoc single-modal branches for each input to leverage its information. Our mutual information based sample acquisition strategy Single-Modal Entropic Measure (SMEM) in addition to our self-distillation technique enables the sample acquisitor to exploit all present modalities and find the most informative samples. Our novel idea is simple to implement, cost-efficient, and readily adaptable to other multi-modal tasks. We confirm our findings on various VQA datasets through state-of-the-art performance by comparing to existing Active Learning baselines.

Dong-Jin Kim, Xiao Sun, Jinsoo Choi et al.

A common problem in the task of human-object interaction (HOI) detection is that numerous HOI classes have only a small number of labeled examples, resulting in training sets with a long-tailed distribution. The lack of positive labels can lead to low classification accuracy for these classes. Towards addressing this issue, we observe that there exist natural correlations and anti-correlations among human-object interactions. In this paper, we model the correlations as action co-occurrence matrices and present techniques to learn these priors and leverage them for more effective training, especially on rare classes. The efficacy of our approach is demonstrated experimentally, where the performance of our approach consistently improves over the state-of-the-art methods on both of the two leading HOI detection benchmark datasets, HICO-Det and V-COCO.

Inkyu Shin, Dong-jin Kim, Jae Won Cho et al.

Unsupervised Domain Adaptation (UDA) for semantic segmentation has been actively studied to mitigate the domain gap between label-rich source data and unlabeled target data. Despite these efforts, UDA still has a long way to go to reach the fully supervised performance. To this end, we propose a Labeling Only if Required strategy, LabOR, where we introduce a human-in-the-loop approach to adaptively give scarce labels to points that a UDA model is uncertain about. In order to find the uncertain points, we generate an inconsistency mask using the proposed adaptive pixel selector and we label these segment-based regions to achieve near supervised performance with only a small fraction (about 2.2%) ground truth points, which we call "Segment based Pixel-Labeling (SPL)". To further reduce the efforts of the human annotator, we also propose "Point-based Pixel-Labeling (PPL)", which finds the most representative points for labeling within the generated inconsistency mask. This reduces efforts from 2.2% segment label to 40 points label while minimizing performance degradation. Through extensive experimentation, we show the advantages of this new framework for domain adaptive semantic segmentation while minimizing human labor costs.

Jae Won Cho, Dong-Jin Kim, Yunjae Jung et al.

Recent state-of-the-art active learning methods have mostly leveraged Generative Adversarial Networks (GAN) for sample acquisition; however, GAN is usually known to suffer from instability and sensitivity to hyper-parameters. In contrast to these methods, we propose in this paper a novel active learning framework that we call Maximum Classifier Discrepancy for Active Learning (MCDAL) which takes the prediction discrepancies between multiple classifiers. In particular, we utilize two auxiliary classification layers that learn tighter decision boundaries by maximizing the discrepancies among them. Intuitively, the discrepancies in the auxiliary classification layers' predictions indicate the uncertainty in the prediction. In this regard, we propose a novel method to leverage the classifier discrepancies for the acquisition function for active learning. We also provide an interpretation of our idea in relation to existing GAN based active learning methods and domain adaptation frameworks. Moreover, we empirically demonstrate the utility of our approach where the performance of our approach exceeds the state-of-the-art methods on several image classification and semantic segmentation datasets in active learning setups.

Youngtaek Oh, Dong-Jin Kim, In So Kweon

The capability of the traditional semi-supervised learning (SSL) methods is far from real-world application due to severely biased pseudo-labels caused by (1) class imbalance and (2) class distribution mismatch between labeled and unlabeled data. This paper addresses such a relatively under-explored problem. First, we propose a general pseudo-labeling framework that class-adaptively blends the semantic pseudo-label from a similarity-based classifier to the linear one from the linear classifier, after making the observation that both types of pseudo-labels have complementary properties in terms of bias. We further introduce a novel semantic alignment loss to establish balanced feature representation to reduce the biased predictions from the classifier. We term the whole framework as Distribution-Aware Semantics-Oriented (DASO) Pseudo-label. We conduct extensive experiments in a wide range of imbalanced benchmarks: CIFAR10/100-LT, STL10-LT, and large-scale long-tailed Semi-Aves with open-set class, and demonstrate that, the proposed DASO framework reliably improves SSL learners with unlabeled data especially when both (1) class imbalance and (2) distribution mismatch dominate.

Jae Won Cho, Dong-Jin Kim, Jinsoo Choi et al.

In this work, we address the issues of missing modalities that have arisen from the Visual Question Answer-Difference prediction task and find a novel method to solve the task at hand. We address the missing modality-the ground truth answers-that are not present at test time and use a privileged knowledge distillation scheme to deal with the issue of the missing modality. In order to efficiently do so, we first introduce a model, the "Big" Teacher, that takes the image/question/answer triplet as its input and outperforms the baseline, then use a combination of models to distill knowledge to a target network (student) that only takes the image/question pair as its inputs. We experiment our models on the VizWiz and VQA-V2 Answer Difference datasets and show through extensive experimentation and ablation the performances of our method and a diverse possibility for future research.

Dong-Jin Kim, Tae-Hyun Oh, Jinsoo Choi et al.

We introduce dense relational captioning, a novel image captioning task which aims to generate multiple captions with respect to relational information between objects in a visual scene. Relational captioning provides explicit descriptions for each relationship between object combinations. This framework is advantageous in both diversity and amount of information, leading to a comprehensive image understanding based on relationships, e.g., relational proposal generation. For relational understanding between objects, the part-of-speech (POS; i.e., subject-object-predicate categories) can be a valuable prior information to guide the causal sequence of words in a caption. We enforce our framework to learn not only to generate captions but also to understand the POS of each word. To this end, we propose the multi-task triple-stream network (MTTSNet) which consists of three recurrent units responsible for each POS which is trained by jointly predicting the correct captions and POS for each word. In addition, we found that the performance of MTTSNet can be improved by modulating the object embeddings with an explicit relational module. We demonstrate that our proposed model can generate more diverse and richer captions, via extensive experimental analysis on large scale datasets and several metrics. Then, we present applications of our framework to holistic image captioning, scene graph generation, and retrieval tasks.

Dong-Jin Kim, Xiao Sun, Jinsoo Choi et al.

A common problem in human-object interaction (HOI) detection task is that numerous HOI classes have only a small number of labeled examples, resulting in training sets with a long-tailed distribution. The lack of positive labels can lead to low classification accuracy for these classes. Towards addressing this issue, we observe that there exist natural correlations and anti-correlations among human-object interactions. In this paper, we model the correlations as action co-occurrence matrices and present techniques to learn these priors and leverage them for more effective training, especially in rare classes. The utility of our approach is demonstrated experimentally, where the performance of our approach exceeds the state-of-the-art methods on both of the two leading HOI detection benchmark datasets, HICO-Det and V-COCO.

Dong-Jin Kim, Jinsoo Choi, Tae-Hyun Oh et al.

Our goal in this work is to train an image captioning model that generates more dense and informative captions. We introduce "relational captioning," a novel image captioning task which aims to generate multiple captions with respect to relational information between objects in an image. Relational captioning is a framework that is advantageous in both diversity and amount of information, leading to image understanding based on relationships. Part-of speech (POS, i.e. subject-object-predicate categories) tags can be assigned to every English word. We leverage the POS as a prior to guide the correct sequence of words in a caption. To this end, we propose a multi-task triple-stream network (MTTSNet) which consists of three recurrent units for the respective POS and jointly performs POS prediction and captioning. We demonstrate more diverse and richer representations generated by the proposed model against several baselines and competing methods.

Dong-Jin Kim, Jinsoo Choi, Tae-Hyun Oh et al.

Human behavior understanding is arguably one of the most important mid-level components in artificial intelligence. In order to efficiently make use of data, multi-task learning has been studied in diverse computer vision tasks including human behavior understanding. However, multi-task learning relies on task specific datasets and constructing such datasets can be cumbersome. It requires huge amounts of data, labeling efforts, statistical consideration etc. In this paper, we leverage existing single-task datasets for human action classification and captioning data for efficient human behavior learning. Since the data in each dataset has respective heterogeneous annotations, traditional multi-task learning is not effective in this scenario. To this end, we propose a novel alternating directional optimization method to efficiently learn from the heterogeneous data. We demonstrate the effectiveness of our model and show performance improvements on both classification and sentence retrieval tasks in comparison to the models trained on each of the single-task datasets.