Satoshi Tsutsui

Satoshi Tsutsui, Zhengyang Su, Bihan Wen

Recognizing the types of white blood cells (WBCs) in microscopic images of human blood smears is a fundamental task in the fields of pathology and hematology. Although previous studies have made significant contributions to the development of methods and datasets, few papers have investigated benchmarks or baselines that others can easily refer to. For instance, we observed notable variations in the reported accuracies of the same Convolutional Neural Network (CNN) model across different studies, yet no public implementation exists to reproduce these results. In this paper, we establish a benchmark for WBC recognition. Our results indicate that CNN-based models achieve high accuracy when trained and tested under similar imaging conditions. However, their performance drops significantly when tested under different conditions. Moreover, the ResNet classifier, which has been widely employed in previous work, exhibits an unreasonably poor generalization ability under domain shifts due to batch normalization. We investigate this issue and suggest some alternative normalization techniques that can mitigate it. We make fully-reproducible code publicly available\footnote{\url{https://github.com/apple2373/wbc-benchmark}}.

Satoshi Tsutsui, Weijia Mao, Sijing Lin et al.

Rendering scenes with a high-quality human face from arbitrary viewpoints is a practical and useful technique for many real-world applications. Recently, Neural Radiance Fields (NeRF), a rendering technique that uses neural networks to approximate classical ray tracing, have been considered as one of the promising approaches for synthesizing novel views from a sparse set of images. We find that NeRF can render new views while maintaining geometric consistency, but it does not properly maintain skin details, such as moles and pores. These details are important particularly for faces because when we look at an image of a face, we are much more sensitive to details than when we look at other objects. On the other hand, 3D Morpable Models (3DMMs) based on traditional meshes and textures can perform well in terms of skin detail despite that it has less precise geometry and cannot cover the head and the entire scene with background. Based on these observations, we propose a method to use both NeRF and 3DMM to synthesize a high-fidelity novel view of a scene with a face. Our method learns a Generative Adversarial Network (GAN) to mix a NeRF-synthesized image and a 3DMM-rendered image and produces a photorealistic scene with a face preserving the skin details. Experiments with various real-world scenes demonstrate the effectiveness of our approach. The code will be available on https://github.com/showlab/headshot .

Satoshi Tsutsui, Yanwei Fu, David Crandall

One-shot fine-grained visual recognition often suffers from the problem of having few training examples for new fine-grained classes. To alleviate this problem, off-the-shelf image generation techniques based on Generative Adversarial Networks (GANs) can potentially create additional training images. However, these GAN-generated images are often not helpful for actually improving the accuracy of one-shot fine-grained recognition. In this paper, we propose a meta-learning framework to combine generated images with original images, so that the resulting "hybrid" training images improve one-shot learning. Specifically, the generic image generator is updated by a few training instances of novel classes, and a Meta Image Reinforcing Network (MetaIRNet) is proposed to conduct one-shot fine-grained recognition as well as image reinforcement. Our experiments demonstrate consistent improvement over baselines on one-shot fine-grained image classification benchmarks. Furthermore, our analysis shows that the reinforced images have more diversity compared to the original and GAN-generated images.

Satoshi Tsutsui, Winnie Pang, Bihan Wen

The examination of blood samples at a microscopic level plays a fundamental role in clinical diagnostics, influencing a wide range of medical conditions. For instance, an in-depth study of White Blood Cells (WBCs), a crucial component of our blood, is essential for diagnosing blood-related diseases such as leukemia and anemia. While multiple datasets containing WBC images have been proposed, they mostly focus on cell categorization, often lacking the necessary morphological details to explain such categorizations, despite the importance of explainable artificial intelligence (XAI) in medical domains. This paper seeks to address this limitation by introducing comprehensive annotations for WBC images. Through collaboration with pathologists, a thorough literature review, and manual inspection of microscopic images, we have identified 11 morphological attributes associated with the cell and its components (nucleus, cytoplasm, and granules). We then annotated ten thousand WBC images with these attributes. Moreover, we conduct experiments to predict these attributes from images, providing insights beyond basic WBC classification. As the first public dataset to offer such extensive annotations, we also illustrate specific applications that can benefit from our attribute annotations. Overall, our dataset paves the way for interpreting WBC recognition models, further advancing XAI in the fields of pathology and hematology.

Xizhe Xue, Dongdong Yu, Lingqiao Liu et al.

Open-World Instance Segmentation (OWIS) is an emerging research topic that aims to segment class-agnostic object instances from images. The mainstream approaches use a two-stage segmentation framework, which first locates the candidate object bounding boxes and then performs instance segmentation. In this work, we instead promote a single-stage framework for OWIS. We argue that the end-to-end training process in the single-stage framework can be more convenient for directly regularizing the localization of class-agnostic object pixels. Based on the single-stage instance segmentation framework, we propose a regularization model to predict foreground pixels and use its relation to instance segmentation to construct a cross-task consistency loss. We show that such a consistency loss could alleviate the problem of incomplete instance annotation -- a common problem in the existing OWIS datasets. We also show that the proposed loss lends itself to an effective solution to semi-supervised OWIS that could be considered an extreme case that all object annotations are absent for some images. Our extensive experiments demonstrate that the proposed method achieves impressive results in both fully-supervised and semi-supervised settings. Compared to SOTA methods, the proposed method significantly improves the $AP_{100}$ score by 4.75\% in UVO$\rightarrow$UVO setting and 4.05\% in COCO$\rightarrow$UVO setting. In the case of semi-supervised learning, our model learned with only 30\% labeled data, even outperforms its fully-supervised counterpart with 50\% labeled data. The code will be released soon.

Satoshi Tsutsui, Xizi Wang, Guangyuan Weng et al.

Machine learning models of visual action recognition are typically trained and tested on data from specific situations where actions are associated with certain objects. It is an open question how action-object associations in the training set influence a model's ability to generalize beyond trained situations. We set out to identify properties of training data that lead to action recognition models with greater generalization ability. To do this, we take inspiration from a cognitive mechanism called cross-situational learning, which states that human learners extract the meaning of concepts by observing instances of the same concept across different situations. We perform controlled experiments with various types of action-object associations, and identify key properties of action-object co-occurrence in training data that lead to better classifiers. Given that these properties are missing in the datasets that are typically used to train action classifiers in the computer vision literature, our work provides useful insights on how we should best construct datasets for efficiently training for better generalization.

Juan Hu, Xin Liao, Difei Gao et al.

Deepfake techniques have been widely used for malicious purposes, prompting extensive research interest in developing Deepfake detection methods. Deepfake manipulations typically involve tampering with facial parts, which can result in inconsistencies across different parts of the face. For instance, Deepfake techniques may change smiling lips to an upset lip, while the eyes remain smiling. Existing detection methods depend on specific indicators of forgery, which tend to disappear as the forgery patterns are improved. To address the limitation, we propose Mover, a new Deepfake detection model that exploits unspecific facial part inconsistencies, which are inevitable weaknesses of Deepfake videos. Mover randomly masks regions of interest (ROIs) and recovers faces to learn unspecific features, which makes it difficult for fake faces to be recovered, while real faces can be easily recovered. Specifically, given a real face image, we first pretrain a masked autoencoder to learn facial part consistency by dividing faces into three parts and randomly masking ROIs, which are then recovered based on the unmasked facial parts. Furthermore, to maximize the discrepancy between real and fake videos, we propose a novel model with dual networks that utilize the pretrained encoder and masked autoencoder, respectively. 1) The pretrained encoder is finetuned for capturing the encoding of inconsistent information in the given video. 2) The pretrained masked autoencoder is utilized for mapping faces and distinguishing real and fake videos. Our extensive experiments on standard benchmarks demonstrate that Mover is highly effective.

Juan Hu, Xin Liao, Difei Gao et al.

The exploitation of Deepfake techniques for malicious intentions has driven significant research interest in Deepfake detection. Deepfake manipulations frequently introduce random tampered traces, leading to unpredictable outcomes in different facial regions. However, existing detection methods heavily rely on specific forgery indicators, and as the forgery mode improves, these traces become increasingly randomized, resulting in a decline in the detection performance of methods reliant on specific forgery traces. To address the limitation, we propose Recap, a novel Deepfake detection model that exposes unspecific facial part inconsistencies by recovering faces and enlarges the differences between real and fake by mapping recovered faces. In the recovering stage, the model focuses on randomly masking regions of interest (ROIs) and reconstructing real faces without unpredictable tampered traces, resulting in a relatively good recovery effect for real faces while a poor recovery effect for fake faces. In the mapping stage, the output of the recovery phase serves as supervision to guide the facial mapping process. This mapping process strategically emphasizes the mapping of fake faces with poor recovery, leading to a further deterioration in their representation, while enhancing and refining the mapping of real faces with good representation. As a result, this approach significantly amplifies the discrepancies between real and fake videos. Our extensive experiments on standard benchmarks demonstrate that Recap is effective in multiple scenarios.

Satoshi Tsutsui, Winnie Pang, Shuting He et al.

The microscopic examination of white blood cells (WBCs) plays a fundamental role in pathology and is essential for diagnosing blood disorders such as leukemia and anemia. To support further research on WBC images, multiple datasets have been proposed. However, they mainly annotate cell categories, and lack detailed morphological characteristics that pathologists use to explain their interpretations of cells. To address this gap, we introduce WBCAtt+, a novel dataset of WBC images densely annotated with 11 morphological attributes and five pixel-level cell components. With 113k image-level labels and 10k segmentation maps, WBCAtt+ is the first to provide comprehensive annotations for WBC images. Leveraging this dataset, we provide baseline models for attribute recognition and semantic segmentation. We also design an attribute recognition model to incorporate compositional structure of cells, further improving the recognition performance. Lastly, we showcase various applications enabled by our dataset, such as explainable AI models, including counterfactual example generation. \revision{The dataset and code are publicly available\footnote{https://doi.org/10.57967/hf/8143}}.

Winnie Pang, Xueyi Ke, Satoshi Tsutsui et al.

Concept bottleneck models (CBMs), which predict human-interpretable concepts (e.g., nucleus shapes in cell images) before predicting the final output (e.g., cell type), provide insights into the decision-making processes of the model. However, training CBMs solely in a data-driven manner can introduce undesirable biases, which may compromise prediction performance, especially when the trained models are evaluated on out-of-domain images (e.g., those acquired using different devices). To mitigate this challenge, we propose integrating clinical knowledge to refine CBMs, better aligning them with clinicians' decision-making processes. Specifically, we guide the model to prioritize the concepts that clinicians also prioritize. We validate our approach on two datasets of medical images: white blood cell and skin images. Empirical validation demonstrates that incorporating medical guidance enhances the model's classification performance on unseen datasets with varying preparation methods, thereby increasing its real-world applicability.

Eric Zhongcong Xu, Zeyang Song, Satoshi Tsutsui et al.

Audio-visual speaker diarization aims at detecting "who spoke when" using both auditory and visual signals. Existing audio-visual diarization datasets are mainly focused on indoor environments like meeting rooms or news studios, which are quite different from in-the-wild videos in many scenarios such as movies, documentaries, and audience sitcoms. To develop diarization methods for these challenging videos, we create the AVA Audio-Visual Diarization (AVA-AVD) dataset. Our experiments demonstrate that adding AVA-AVD into training set can produce significantly better diarization models for in-the-wild videos despite that the data is relatively small. Moreover, this benchmark is challenging due to the diverse scenes, complicated acoustic conditions, and completely off-screen speakers. As a first step towards addressing the challenges, we design the Audio-Visual Relation Network (AVR-Net) which introduces a simple yet effective modality mask to capture discriminative information based on face visibility. Experiments show that our method not only can outperform state-of-the-art methods but is more robust as varying the ratio of off-screen speakers. Our data and code has been made publicly available at https://github.com/showlab/AVA-AVD.

Xiyu Wang, Yufei Wang, Satoshi Tsutsui et al.

Diffusion-based models for story visualization have shown promise in generating content-coherent images for storytelling tasks. However, how to effectively integrate new characters into existing narratives while maintaining character consistency remains an open problem, particularly with limited data. Two major limitations hinder the progress: (1) the absence of a suitable benchmark due to potential character leakage and inconsistent text labeling, and (2) the challenge of distinguishing between new and old characters, leading to ambiguous results. To address these challenges, we introduce the NewEpisode benchmark, comprising refined datasets designed to evaluate generative models' adaptability in generating new stories with fresh characters using just a single example story. The refined dataset involves refined text prompts and eliminates character leakage. Additionally, to mitigate the character confusion of generated results, we propose EpicEvo, a method that customizes a diffusion-based visual story generation model with a single story featuring the new characters seamlessly integrating them into established character dynamics. EpicEvo introduces a novel adversarial character alignment module to align the generated images progressively in the diffusive process, with exemplar images of new characters, while applying knowledge distillation to prevent forgetting of characters and background details. Our evaluation quantitatively demonstrates that EpicEvo outperforms existing baselines on the NewEpisode benchmark, and qualitative studies confirm its superior customization of visual story generation in diffusion models. In summary, EpicEvo provides an effective way to incorporate new characters using only one example story, unlocking new possibilities for applications such as serialized cartoons.

Ziwang Xu, Lanqing Guo, Satoshi Tsutsui et al.

Staining is essential in cell imaging and medical diagnostics but poses significant challenges, including high cost, time consumption, labor intensity, and irreversible tissue alterations. Recent advances in deep learning have enabled digital staining through supervised model training. However, collecting large-scale, perfectly aligned pairs of stained and unstained images remains difficult. In this work, we propose a novel unsupervised deep learning framework for digital cell staining that reduces the need for extensive paired data using knowledge distillation. We explore two training schemes: (1) unpaired and (2) paired-but-misaligned settings. For the unpaired case, we introduce a two-stage pipeline, comprising light enhancement followed by colorization, as a teacher model. Subsequently, we obtain a student staining generator through knowledge distillation with hybrid non-reference losses. To leverage the pixel-wise information between adjacent sections, we further extend to the paired-but-misaligned setting, adding the Learning to Align module to utilize pixel-level information. Experiment results on our dataset demonstrate that our proposed unsupervised deep staining method can generate stained images with more accurate positions and shapes of the cell targets in both settings. Compared with competing methods, our method achieves improved results both qualitatively and quantitatively (e.g., NIQE and PSNR).We applied our digital staining method to the White Blood Cell (WBC) dataset, investigating its potential for medical applications.

Xueyi Ke, Satoshi Tsutsui, Yayun Zhang et al.

Infants develop complex visual understanding rapidly, even preceding the acquisition of linguistic skills. As computer vision seeks to replicate the human vision system, understanding infant visual development may offer valuable insights. In this paper, we present an interdisciplinary study exploring this question: can a computational model that imitates the infant learning process develop broader visual concepts that extend beyond the vocabulary it has heard, similar to how infants naturally learn? To investigate this, we analyze a recently published model in Science by Vong et al., which is trained on longitudinal, egocentric images of a single child paired with transcribed parental speech. We perform neuron labeling to identify visual concept neurons hidden in the model's internal representations. We then demonstrate that these neurons can recognize objects beyond the model's original vocabulary. Furthermore, we compare the differences in representation between infant models and those in modern computer vision models, such as CLIP and ImageNet pre-trained model. Ultimately, our work bridges cognitive science and computer vision by analyzing the internal representations of a computational model trained on an infant visual and linguistic inputs. Project page is available at https://kexueyi.github.io/webpage-discover-hidden-visual-concepts.

Yuxuan Cai, Xiyu Wang, Satoshi Tsutsui et al.

Concept Bottleneck Models (CBMs) aim to enhance interpretability by predicting human-understandable concepts as intermediates for decision-making. However, these models often face challenges in ensuring reliable concept representations, which can propagate to downstream tasks and undermine robustness, especially under distribution shifts. Two inherent issues contribute to concept unreliability: sensitivity to concept-irrelevant features (e.g., background variations) and lack of semantic consistency for the same concept across different samples. To address these limitations, we propose the Reliability-Enhanced Concept Embedding Model (RECEM), which introduces a two-fold strategy: Concept-Level Disentanglement to separate irrelevant features from concept-relevant information and a Concept Mixup mechanism to ensure semantic alignment across samples. These mechanisms work together to improve concept reliability, enabling the model to focus on meaningful object attributes and generate faithful concept representations. Experimental results demonstrate that RECEM consistently outperforms existing baselines across multiple datasets, showing superior performance under background and domain shifts. These findings highlight the effectiveness of disentanglement and alignment strategies in enhancing both reliability and robustness in CBMs.

Juan Hu, Xin Liao, Difei Gao et al.

Deepfake videos are becoming increasingly realistic, showing few tampering traces on facial areasthat vary between frames. Consequently, existing Deepfake detection methods struggle to detect unknown domain Deepfake videos while accurately locating the tampered region. To address thislimitation, we propose Delocate, a novel Deepfake detection model that can both recognize andlocalize unknown domain Deepfake videos. Ourmethod consists of two stages named recoveringand localization. In the recovering stage, the modelrandomly masks regions of interest (ROIs) and reconstructs real faces without tampering traces, leading to a relatively good recovery effect for realfaces and a poor recovery effect for fake faces. Inthe localization stage, the output of the recoveryphase and the forgery ground truth mask serve assupervision to guide the forgery localization process. This process strategically emphasizes the recovery phase of fake faces with poor recovery, facilitating the localization of tampered regions. Ourextensive experiments on four widely used benchmark datasets demonstrate that Delocate not onlyexcels in localizing tampered areas but also enhances cross-domain detection performance.

Satoshi Tsutsui, Ruta Desai, Karl Ridgeway

Understanding users' activities from head-mounted cameras is a fundamental task for Augmented and Virtual Reality (AR/VR) applications. A typical approach is to train a classifier in a supervised manner using data labeled by humans. This approach has limitations due to the expensive annotation cost and the closed coverage of activity labels. A potential way to address these limitations is to use self-supervised learning (SSL). Instead of relying on human annotations, SSL leverages intrinsic properties of data to learn representations. We are particularly interested in learning egocentric video representations benefiting from the head-motion generated by users' daily activities, which can be easily obtained from IMU sensors embedded in AR/VR devices. Towards this goal, we propose a simple but effective approach to learn video representation by learning to tell the corresponding pairs of video clip and head-motion. We demonstrate the effectiveness of our learned representation for recognizing egocentric activities of people and dogs.

Satoshi Tsutsui, David Crandall, Chen Yu

We analyze egocentric views of attended objects from infants. This paper shows 1) empirical evidence that children's egocentric views have more diverse distributions compared to adults' views, 2) we can computationally simulate the infants' distribution, and 3) the distribution is beneficial for training more generalized image classifiers not only for infant egocentric vision but for third-person computer vision.

Satoshi Tsutsui, Yanwei Fu, David Crandall

Recognizing people by faces and other biometrics has been extensively studied in computer vision. But these techniques do not work for identifying the wearer of an egocentric (first-person) camera because that person rarely (if ever) appears in their own first-person view. But while one's own face is not frequently visible, their hands are: in fact, hands are among the most common objects in one's own field of view. It is thus natural to ask whether the appearance and motion patterns of people's hands are distinctive enough to recognize them. In this paper, we systematically study the possibility of Egocentric Hand Identification (EHI) with unconstrained egocentric hand gestures. We explore several different visual cues, including color, shape, skin texture, and depth maps to identify users' hands. Extensive ablation experiments are conducted to analyze the properties of hands that are most distinctive. Finally, we show that EHI can improve generalization of other tasks, such as gesture recognition, by training adversarially to encourage these models to ignore differences between users.

Satoshi Tsutsui, Arjun Chandrasekaran, Md Alimoor Reza et al.

Human infants have the remarkable ability to learn the associations between object names and visual objects from inherently ambiguous experiences. Researchers in cognitive science and developmental psychology have built formal models that implement in-principle learning algorithms, and then used pre-selected and pre-cleaned datasets to test the abilities of the models to find statistical regularities in the input data. In contrast to previous modeling approaches, the present study used egocentric video and gaze data collected from infant learners during natural toy play with their parents. This allowed us to capture the learning environment from the perspective of the learner's own point of view. We then used a Convolutional Neural Network (CNN) model to process sensory data from the infant's point of view and learn name-object associations from scratch. As the first model that takes raw egocentric video to simulate infant word learning, the present study provides a proof of principle that the problem of early word learning can be solved, using actual visual data perceived by infant learners. Moreover, we conducted simulation experiments to systematically determine how visual, perceptual, and attentional properties of infants' sensory experiences may affect word learning.

Satoshi Tsutsui, Yanwei Fu, David Crandall

One-shot fine-grained visual recognition often suffers from the problem of training data scarcity for new fine-grained classes. To alleviate this problem, an off-the-shelf image generator can be applied to synthesize additional training images, but these synthesized images are often not helpful for actually improving the accuracy of one-shot fine-grained recognition. This paper proposes a meta-learning framework to combine generated images with original images, so that the resulting ``hybrid'' training images can improve one-shot learning. Specifically, the generic image generator is updated by a few training instances of novel classes, and a Meta Image Reinforcing Network (MetaIRNet) is proposed to conduct one-shot fine-grained recognition as well as image reinforcement. The model is trained in an end-to-end manner, and our experiments demonstrate consistent improvement over baselines on one-shot fine-grained image classification benchmarks.

Satoshi Tsutsui, Dian Zhi, Md Alimoor Reza et al.

Inspired by the remarkable ability of the infant visual learning system, a recent study collected first-person images from children to analyze the `training data' that they receive. We conduct a follow-up study that investigates two additional directions. First, given that infants can quickly learn to recognize a new object without much supervision (i.e. few-shot learning), we limit the number of training images. Second, we investigate how children control the supervision signals they receive during learning based on hand manipulation of objects. Our experimental results suggest that supervision with hand manipulation is better than without hands, and the trend is consistent even when a small number of images is available.

Zheng Gao, Gang Fu, Chunping Ouyang et al.

Representation learning provides new and powerful graph analytical approaches and tools for the highly valued data science challenge of mining knowledge graphs. Since previous graph analytical methods have mostly focused on homogeneous graphs, an important current challenge is extending this methodology for richly heterogeneous graphs and knowledge domains. The biomedical sciences are such a domain, reflecting the complexity of biology, with entities such as genes, proteins, drugs, diseases, and phenotypes, and relationships such as gene co-expression, biochemical regulation, and biomolecular inhibition or activation. Therefore, the semantics of edges and nodes are critical for representation learning and knowledge discovery in real world biomedical problems. In this paper, we propose the edge2vec model, which represents graphs considering edge semantics. An edge-type transition matrix is trained by an Expectation-Maximization approach, and a stochastic gradient descent model is employed to learn node embedding on a heterogeneous graph via the trained transition matrix. edge2vec is validated on three biomedical domain tasks: biomedical entity classification, compound-gene bioactivity prediction, and biomedical information retrieval. Results show that by considering edge-types into node embedding learning in heterogeneous graphs, \textbf{edge2vec}\ significantly outperforms state-of-the-art models on all three tasks. We propose this method for its added value relative to existing graph analytical methodology, and in the real world context of biomedical knowledge discovery applicability.

Ting-Ting Liang, Satoshi Tsutsui, Liangcai Gao et al.

One of the time-consuming routine work for a radiologist is to discern anatomical structures from tomographic images. For assisting radiologists, this paper develops an automatic segmentation method for pelvic magnetic resonance (MR) images. The task has three major challenges 1) A pelvic organ can have various sizes and shapes depending on the axial image, which requires local contexts to segment correctly. 2) Different organs often have quite similar appearance in MR images, which requires global context to segment. 3) The number of available annotated images are very small to use the latest segmentation algorithms. To address the challenges, we propose a novel convolutional neural network called Attention-Pyramid network (APNet) that effectively exploits both local and global contexts, in addition to a data-augmentation technique that is particularly effective for MR images. In order to evaluate our method, we construct fine-grained (50 pelvic organs) MR image segmentation dataset, and experimentally confirm the superior performance of our techniques over the state-of-the-art image segmentation methods.

Satoshi Tsutsui, Tommi Kerola, Shunta Saito et al.

Identifying "free-space," or safely driveable regions in the scene ahead, is a fundamental task for autonomous navigation. While this task can be addressed using semantic segmentation, the manual labor involved in creating pixelwise annotations to train the segmentation model is very costly. Although weakly supervised segmentation addresses this issue, most methods are not designed for free-space. In this paper, we observe that homogeneous texture and location are two key characteristics of free-space, and develop a novel, practical framework for free-space segmentation with minimal human supervision. Our experiments show that our framework performs better than other weakly supervised methods while using less supervision. Our work demonstrates the potential for performing free-space segmentation without tedious and costly manual annotation, which will be important for adapting autonomous driving systems to different types of vehicles and environments

Satoshi Tsutsui, Tommi Kerola, Shunta Saito

We present an approach for road segmentation that only requires image-level annotations at training time. We leverage distant supervision, which allows us to train our model using images that are different from the target domain. Using large publicly available image databases as distant supervisors, we develop a simple method to automatically generate weak pixel-wise road masks. These are used to iteratively train a fully convolutional neural network, which produces our final segmentation model. We evaluate our method on the Cityscapes dataset, where we compare it with a fully supervised approach. Further, we discuss the trade-off between annotation cost and performance. Overall, our distantly supervised approach achieves 93.8% of the performance of the fully supervised approach, while using orders of magnitude less annotation work.

Satoshi Tsutsui, David Crandall

Recent work in computer vision has yielded impressive results in automatically describing images with natural language. Most of these systems generate captions in a sin- gle language, requiring multiple language-specific models to build a multilingual captioning system. We propose a very simple technique to build a single unified model across languages, using artificial tokens to control the language, making the captioning system more compact. We evaluate our approach on generating English and Japanese captions, and show that a typical neural captioning architecture is capable of learning a single model that can switch between two different languages.

Satoshi Tsutsui, David Crandall

A key problem in automatic analysis and understanding of scientific papers is to extract semantic information from non-textual paper components like figures, diagrams, tables, etc. Much of this work requires a very first preprocessing step: decomposing compound multi-part figures into individual subfigures. Previous work in compound figure separation has been based on manually designed features and separation rules, which often fail for less common figure types and layouts. Moreover, few implementations for compound figure decomposition are publicly available. This paper proposes a data driven approach to separate compound figures using modern deep Convolutional Neural Networks (CNNs) to train the separator in an end-to-end manner. CNNs eliminate the need for manually designing features and separation rules, but require a large amount of annotated training data. We overcome this challenge using transfer learning as well as automatically synthesizing training exemplars. We evaluate our technique on the ImageCLEF Medical dataset, achieving 85.9% accuracy and outperforming previous techniques. We have released our implementation as an easy-to-use Python library, aiming to promote further research in scientific figure mining.