arXiv subjects

Hirokatsu Kataoka

Publications and source records attributed to Hirokatsu Kataoka.

At least 19 recordsLinked to original sources

arXiv2026

Improving Imitation Learning Efficiency for Manipulation through Geometric Prior Pretraining

Applying an imitation learning policy to a new manipulation task usually requires collecting new demonstrations and retraining the model, which makes sample efficiency a practical concern. Pretraining on large-scale robot datasets is effective in this respect, but such datasets are costly to collect and train on, while data augmentation techniques typically require a new round of data generation and retraining for each task. A complementary question is what useful prior can be provided to a policy at negligible cost before any task-specific data are collected. In this study, we construct a geometric visual pretraining dataset in which each scene contains only a plane, an object, and a hand, and trajectories are generated automatically. The scenes contain neither textures nor backgrounds; pretraining primarily exposes the policy to the geometric relationship between the hand and the object. Furthermore, representing the hand as a cube avoids tailoring the dataset to a specific robot morphology. We evaluate this geometric prior using ACT on three simulated robots across five manipulation tasks each, as well as on three real-world robot tasks. Across many of these robot--task combinations, fine-tuning from the geometric prior achieves higher success rates in the early stages of training than training from scratch while using only a small number of task demonstrations. These results suggest that even highly simplified geometric scenes can provide a useful initialization that transfers across robots and to real-world tasks when task data are limited.

cs.RO↗

arXiv2026

Visual General Intelligence: A White Paper

This paper reconsiders intelligence from a vision-centered perspective and examines whether intelligence emerging from visual experience and learning may provide a pathway toward AGI. In the language domain, beginning with the introduction of the Transformer architecture, the GPT series has demonstrated transfer to unseen tasks through autoregressive language modeling on web-scale text combined with aggressive scaling. This raises a natural question, namely, what capabilities and forms of intelligence can emerge from visual modalities such as images, videos, and geometry? In this paper, we discuss whether visual intelligence can serve as a pathway toward AGI, referred to in this paper as visual general intelligence (VGI), by bringing together contributors from diverse standpoints and affiliations. Our aim is not to offer a single definition of visual intelligence, but to clarify the principles that computer vision should pursue in the AGI era, the visual input modalities, the benchmarks, the learning paradigms, and the relationship between vision, when taken as the core, and other modalities such as language.

cs.CV↗

arXiv2026

Beyond Single Object: Learning 3D Relations with Large Language Models

We address a fundamental gap in 3D-LLMs: existing models focus on single-object/scene description, struggling with detailed, inter-object comparison. We propose a framework for detailed object-level reasoning across multiple objects with three components: (1) MO3D (Multi-Object in 3D), an instruction dataset requiring fine-grained multi-object comparison; (2) Multi-3DLLM, using a minimal Patch-Interaction Transformer (PIT) that models inter-/intra-object relationships while preserving local geometry; (3) Mini-apps, two application-driven benchmarks (Shape Mating, Change Captioning) that probe geometric understanding for practical use. Recent 3D-LLMs and 2D-VLMs perform poorly on these tasks, lacking both comparison-centric design and geometric awareness. In contrast, Multi-3DLLM trained on our mixture data learns geometric reasoning, surpasses all baselines on MO3D, and provides positive transfer to single-object classification.

cs.CV↗

arXiv2026

Seeing Red, Thinking Bad: Color Bias in Vision Language Models

Vision language models (VLMs) are increasingly used in industrial decision-making systems, such as recruitment support and recommendation. This motivates careful analysis of how VLMs process visual and textual information. In this work, we study how VLMs interpret text rendered as an image, and investigate the influence of visual styling biases. To this end, we introduce Stealth Visual Prompts, which subtly change visual styling of text, such as color and contrast, while preserving semantic content. Using these prompts, we systematically control the visual styling of words in text and measure their impact on the analysis performed by VLMs. We further analyze how such visual perturbations affect the latent representations of the vision encoder. From our experiments, we observed that coloring positive words in green consistently shifts sentiment predictions toward a positive direction. As a result, VLMs often fail to properly account for negative words present in the text. Our analysis suggests that this behavior is correlated with changes in the latent representations of the vision encoder induced by color variations. In addition, we show that reducing text--background contrast increases reliance on visually salient cues and leads to more incorrect Visual Question Answering (VQA) outputs. These results suggest that the visual styling of rendered text can guide VLMs' interpretation in ways that diverge from human semantic understanding. Project page: https://github.com/KohsukeIde/color-bias-vlm

cs.CV↗

arXiv2026

Industrial Synthetic Segment Pre-training

Vision Foundation Models (VFMs) have made remarkable progress and are increasingly being applied to segmentation tasks in real-world industrial settings. However, VFMs pre-trained on real-image datasets still face several challenges: (1) they do not always perform well on industrial datasets due to significant differences from natural imagery, (2) legal and ethical restrictions, such as limitations on commercial use, constrain extensibility, and (3) building training frameworks under limited computational and data resources remains a critical issue. These challenges raise a fundamental question: can we construct industrial segmentation models without relying on real images or manual annotations? To address this question, we propose the Instance Core Segment Dataset (InsCore), a synthetic data generation framework and the resulting pre-training dataset based on Formula-Driven Supervised Learning (FDSL). InsCore is designed not around the visual appearance or domain of real images, but around the hypothesis that learning to handle complex occlusions during pre-training is a key factor for strong performance in industrial domains. Through experiments across five domains (medical, biomedical, remote sensing, manufacturing, and logistics) we demonstrate that InsCore pre-trained models achieve average mAP scores of 45.2 with the ViTDet backbone and 46.0 with the Swin Transformer backbone, on par with ImageNet-21k supervised pre-training (45.0) while using no real images at all. As a reference point under different input assumptions, prompted SAM with ground-truth bounding boxes attains 45.4 on the same benchmarks. Finally, InsCore consists of only 100k images and 3.2M masks, roughly 1/110 and 1/312 the scale of the SA-1B dataset.

cs.CV↗

arXiv2026

Image-Space Rule Discovery

Can image-editing models discover visual rules in image space and complete problem-solving end-to-end? We tackle this question in the spirit of a human worksheet test (e.g., an IQ test), using problems that require models to read image-based instructions, recognize the problem, infer the answer, bind it to the correct destination, control output count, suppress unnecessary edits, and preserve the input and format. We introduce WISRD, a Worksheet Image-Space Rule Discovery benchmark with 11 core tasks under eight information conditions, spanning localized marking, filling, copying, counting, and no-edit suppression, together with four supplementary reasoning-stress probes for multi-step spatial manipulation, abstract pattern reasoning, logical inference, and constraint-based problem solving. We identify three key findings as follows. (i) Among the frontier image-editing models evaluated, Nano Banana Pro achieves the highest score. On the shared V0--V3 no-reference subset, the Auto-Strict proxy pass rates are 48.7% for Nano Banana Pro, 13.4\% for Qwen-Image-Edit, 11.5% for FLUX.2 Klein 4B API, 11.3% for FLUX.2 Klein 4B open-weight, and 0.0% for InstructPix2Pix. (ii) Analysis reveals that current image-editing models can partially rely on rendered in-image instructions even when the external prompt is absent or merely generic. (iii) In small supplementary diagnostics, Nano Banana Pro achieves 70.0% on 4-by-4 Sudoku and 22.9% on public RAVEN pattern-discovery items in image space.

cs.CV↗

arXiv2026

Local Brushstroke Quality Assessment via Vision-Language Feedback

This paper investigates whether multimodal LLMs can evaluate local brushstroke quality in calligraphy and generate educationally useful natural language feedback. We construct an evaluation framework in which three multimodal LLMs (GPT-4o, Claude Sonnet 4, and Gemini 2.5 Flash) assess before-after image pairs of calligraphic works using a five-point ordinal scale, and compare their outputs against scores assigned by three expert calligraphers. We additionally examine a Retrieval-Augmented Generation (RAG) variant of Claude as a preliminary condition. Results show that all models achieve useful levels of absolute score accuracy (MAE), with GPT-4o performing best (MAE = 0.885). However, none of the models produce statistically significant overall rank correlations with human experts (Kendall's tau). Vocabulary analysis of generated rationales reveals characteristic evaluative biases in each model, and RAG is shown to improve rank correlation while worsening absolute accuracy, constituting an important negative result for text-based rule injection.

cs.CV↗

arXiv2026

Feed-forward Motion In-betweening for Any 4D

4D dynamics (3D geometry evolving over time) is a fundamental representation of the physical world and plays a crucial role in world modeling (e.g., animation and games). Owing to the scarcity of large-scale, long-horizon 4D mesh data with arbitrary shapes, early text-to-4D methods rely on distillation or test-time optimization from video diffusion priors, making inference prohibitively slow. Recent feed-forward generators greatly reduce inference cost but offer limited spatiotemporal controllability, and short-horizon generation often leads to error accumulation in long-horizon sequences. We propose a novel feed-forward in-betweening framework for arbitrary 4D meshes with keyframe conditioning. Building on universal mesh-animation latents, we introduce a frame-wise mesh VAE that encodes each frame into topology-agnostic latent tokens anchored by a reference mesh for keyframe conditioning. We further introduce a keyframe-conditioned rectified flow model with an MMDiT backbone that synthesizes non-keyframe frames conditioned on sparse keyframes. Experiments show strong performance and improved controllability on both DyMesh16 and DyMesh32 benchmarks.

cs.CV↗

arXiv2026

S3OD: Towards Generalizable Salient Object Detection with Synthetic Data

Salient object detection exemplifies data-bounded tasks where expensive pixel-precise annotations force separate model training for related subtasks like DIS and HR-SOD. We present a method that dramatically improves generalization through large-scale synthetic data generation and ambiguity-aware architecture. We introduce S3OD, a dataset of over 139,000 high-resolution images created through our multi-modal diffusion pipeline that extracts labels from diffusion and DINO-v3 features. The iterative generation framework prioritizes challenging categories based on model performance. We propose a streamlined multi-mask decoder that handles the inherent ambiguity in salient object detection by predicting multiple valid interpretations. Models trained only on synthetic data achieve 20-50% error reduction in cross-dataset generalization, while fine-tuned versions reach state-of-the-art performance across DIS and HR-SOD benchmarks.

cs.CV↗

arXiv2026

HanDyVQA: A Video QA Benchmark for Fine-Grained Hand-Object Interaction Dynamics

Hand-object interaction (HOI) inherently involves dynamics where human manipulations produce distinct spatio-temporal effects on objects. However, existing semantic HOI benchmarks focused either on manipulation or on the resulting effects at a coarse level, lacking fine-grained spatio-temporal reasoning to capture the underlying dynamics in HOI. We introduce HanDyVQA, a fine-grained video question-answering benchmark that comprehensively covers both the manipulation and effect aspects of HOI. HanDyVQA comprises six complementary question types (Action, Process, Objects, Location, State Change, and Object Parts), totalling 11.1K multiple-choice QA pairs. Collected QA pairs recognizing manipulation styles, hand/object motions, and part-level state changes. HanDyVQA also includes 10.3K segmentation masks for Objects and Object Parts questions, enabling the evaluation of object/part-level reasoning in video object segmentation. We evaluated recent video foundation models on our benchmark and found that even the best-performing model, Gemini-2.5-Pro, reached only 73% average accuracy, which is far from human performance (97%). Further analysis shows the remaining challenges in spatial relationship, motion, and part-level geometric understanding. We also found that integrating explicit HOI-related cues into visual features improves performance, offering insights for developing future models with a deeper understanding of HOI dynamics.

cs.CV↗

arXiv2026

CLIP-like Model as a Foundational Density Ratio Estimator

Density ratio estimation is a core concept in statistical machine learning because it provides a unified mechanism for tasks such as importance weighting, divergence estimation, and likelihood-free inference, but its potential in vision and language models has not been fully explored. Modern vision-language encoders such as CLIP and SigLIP are trained with contrastive objectives that implicitly optimize log density ratios between joint and marginal image-text distributions, which implicitly learn similarity scores proportional to log density ratios. However, prior work has largely focused on their embedding utility, and the density-ratio structure induced by contrastive learning has not been systematically examined or exploited in multimodal applications. To address this gap, we reinterpret CLIP-style models as pretrained and general-purpose density ratio estimators and show that this perspective enables new algorithmic capabilities. We present a unified explanation of how contrastive objectives estimate density ratios and propose two practical applications: Importance Weight Learning and KL divergence estimation. Our Importance Weight Learning method requires only a single additional prompt and improves F1 scores by up to 7 points. We further show that CLIP-based density ratios support estimation of KL divergences that quantify how conditioning on an image or text alters the distribution of the other modality. Through qualitative examples and an N-gram analysis of captions, we find that these divergences capture semantic diversity and mode structure in multimodal data. Leveraging this property, we introduce a simple KL-guided data curation method that achieves performance competitive with LAION2B filtering.

cs.CV↗

arXiv2026

PowerCLIP: Powerset Alignment for Contrastive Pre-Training

Contrastive vision-language pre-training frameworks such as CLIP have demonstrated impressive zero-shot performance across a range of vision-language tasks. Recent studies have shown that aligning individual text tokens with specific image patches or regions enhances fine-grained compositional understanding. However, it remains challenging to capture compositional semantics that span multiple image regions. To address this limitation, we propose PowerCLIP, a novel contrastive pre-training framework enhanced by powerset alignment, which exhaustively optimizes region-to-phrase alignments by minimizing the loss defined between powersets of image regions and textual parse trees. Since the naive powerset construction incurs exponential computational cost due to the combinatorial explosion in the number of region subsets, we introduce efficient non-linear aggregators (NLAs) that reduce complexity from O(2^M) to O(M) with respect to the number of regions M, while approximating the exact loss value with arbitrary precision. Our extensive experiments demonstrate that PowerCLIP outperforms state-of-the-art methods in zero-shot classification and retrieval tasks, underscoring the compositionality and robustness of our approach. Code is available at https://github.com/Masakichi210/PowerCLIP.

cs.CV↗

arXiv2026

FDIF: Formula-Driven supervised Learning with Implicit Functions for 3D Medical Image Segmentation

Deep learning-based 3D medical image segmentation methods relies on large-scale labeled datasets, yet acquiring such data is difficult due to privacy constraints and the high cost of expert annotation. Formula-Driven Supervised Learning (FDSL) offers an appealing alternative by generating training data and labels directly from mathematical formulas. However, existing voxel-based approaches are limited in geometric expressiveness and cannot synthesize realistic textures. We introduce Formula-Driven supervised learning with Implicit Functions (FDIF), a framework that enables scalable pre-training without using any real data and medical expert annotations. FDIF introduces an implicit-function representation based on signed distance functions (SDFs), enabling compact modeling of complex geometries while exploiting the surface representation of SDFs to support controllable synthesis of both geometric and intensity textures. Across three medical image segmentation benchmarks (AMOS, ACDC, and KiTS) and three architectures (SwinUNETR, nnUNet ResEnc-L, and nnUNet Primus-M), FDIF consistently improves over a formula-driven method, and achieves performance comparable to self-supervised approaches pre-trained on large-scale real datasets. We further show that FDIF pre-training also benefits 3D classification tasks, highlighting implicit-function-based formula supervision as a promising paradigm for data-free representation learning. Code is available at https://github.com/yamanoko/FDIF.

cs.CV↗

arXiv2026

3D sans 3D Scans: Scalable Pre-training from Video-Generated Point Clouds

Despite recent progress in 3D self-supervised learning, collecting large-scale 3D scene scans remains expensive and labor-intensive. In this work, we investigate whether 3D representations can be learned from unlabeled videos recorded without any real 3D sensors. We present Laplacian-Aware Multi-level 3D Clustering with Sinkhorn-Knopp (LAM3C), a self-supervised framework that learns from video-generated point clouds reconstructed from unlabeled videos. We first introduce RoomTours, a video-generated point cloud dataset constructed by collecting room-walkthrough videos from the web (e.g., real-estate tours) and generating 49,219 scenes using an off-the-shelf feed-forward reconstruction model. We also propose a noise-regularized loss that stabilizes representation learning by enforcing local geometric smoothness and ensuring feature stability under noisy point clouds. Remarkably, without using any real 3D scans, LAM3C achieves better performance than previous self-supervised methods on indoor semantic and instance segmentation. These results suggest that unlabeled videos represent an abundant source of data for 3D self-supervised learning. Our source code is available at https://ryosuke-yamada.github.io/lam3c/.

cs.CV↗

arXiv2026

MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness

Data augmentation is a key technique for improving the robustness of image classification models. However, many recent approaches rely on diffusion-based synthesis or complex feature mixing strategies, which introduce substantial computational overhead or require external datasets. In this work, we explore a different direction: procedural augmentation based on analytic interference patterns. Unlike conventional augmentation methods that rely on stochastic noise, feature mixing, or generative models, our approach exploits Moire interference to generate structured perturbations spanning a wide range of spatial frequencies. We propose a lightweight augmentation method that procedurally generates Moire textures on-the-fly using a closed-form mathematical formulation. The patterns are synthesized directly in memory with negligible computational cost (0.0026 seconds per image), mixed with training images during training, and immediately discarded, enabling a storage-free augmentation pipeline without external data. Extensive experiments with Vision Transformers demonstrate that the proposed method consistently improves robustness across multiple benchmarks, including ImageNet-C, ImageNet-R, and adversarial benchmarks, outperforming standard augmentation baselines and existing external-data-free augmentation approaches. These results suggest that analytic interference patterns provide a practical and efficient alternative to data-driven generative augmentation methods.

cs.CV↗

arXiv2026

Constant Rate Scheduling: A General Framework for Optimizing Diffusion Noise Schedule via Distributional Change

We propose a general framework for optimizing noise schedules in diffusion models, applicable to both training and sampling. Our method enforces a constant rate of change in the probability distribution of diffused data throughout the diffusion process, where the rate of change is quantified using a user-defined discrepancy measure. We introduce three such measures, which can be flexibly selected or combined depending on the domain and model architecture. While our framework is inspired by theoretical insights, we do not aim to provide a complete theoretical justification of how distributional change affects sample quality. Instead, we focus on establishing a general-purpose scheduling framework and validating its empirical effectiveness. Through extensive experiments, we demonstrate that our approach consistently improves the performance of both pixel-space and latent-space diffusion models, across various datasets, samplers, and a wide range of number of function evaluations from 5 to 250. In particular, when applied to both training and sampling schedules, our method achieves a state-of-the-art FID score of 2.03 on LSUN Horse 256$\times$256, without compromising mode coverage.

cs.CV↗

arXiv2025

Pre-training Vision Transformers with Formula-driven Supervised Learning

In the present work, we show that the performance of formula-driven supervised learning (FDSL) can match or even exceed that of ImageNet-21k and can approach that of the JFT-300M dataset without the use of real images, human supervision, or self-supervision during the pre-training of vision transformers (ViTs). For example, ViT-Base pre-trained on ImageNet-21k and JFT-300M showed 83.0 and 84.1% top-1 accuracy when fine-tuned on ImageNet-1k, and FDSL showed 83.8% top-1 accuracy when pre-trained under comparable conditions (hyperparameters and number of epochs). Especially, the ExFractalDB-21k pre-training was calculated with x14.2 fewer images compared with JFT-300M. Images generated by formulas avoid privacy and copyright issues, labeling costs and errors, and biases that real images suffer from, and thus have tremendous potential for pre-training general models. To understand the performance of the synthetic images, we tested two hypotheses, namely (i) object contours are what matter in FDSL datasets and (ii) an increased number of parameters for label creation improves performance in FDSL pre-training. To test the former hypothesis, we constructed a dataset that consisted of simple object contour combinations. We found that this dataset matched the performance of fractal databases. For the latter hypothesis, we found that increasing the difficulty of the pre-training task generally leads to better fine-tuning accuracy.

cs.CV↗

arXiv2025

Approximate Domain Unlearning for Vision-Language Models

Pre-trained Vision-Language Models (VLMs) exhibit strong generalization capabilities, enabling them to recognize a wide range of objects across diverse domains without additional training. However, they often retain irrelevant information beyond the requirements of specific downstream tasks, raising concerns about computational efficiency and potential information leakage. This has motivated growing interest in approximate unlearning, which aims to selectively remove unnecessary knowledge while preserving overall model performance. Existing approaches to approximate unlearning have primarily focused on class unlearning, where a VLM is retrained to fail to recognize specified object classes while maintaining accuracy for others. However, merely forgetting object classes is often insufficient in practical applications. For instance, an autonomous driving system should accurately recognize real cars while avoiding misrecognition of illustrated cars depicted in roadside advertisements as real cars, which could be hazardous. In this paper, we introduce Approximate Domain Unlearning (ADU), a novel problem setting that requires reducing recognition accuracy for images from specified domains (e.g., illustration) while preserving accuracy for other domains (e.g., real). ADU presents new technical challenges: due to the strong domain generalization capability of pre-trained VLMs, domain distributions are highly entangled in the feature space, making naive approaches based on penalizing target domains ineffective. To tackle this limitation, we propose a novel approach that explicitly disentangles domain distributions and adaptively captures instance-specific domain information. Extensive experiments show that our approach outperforms baselines built upon VLM tuning techniques, paving the way for practical and fine-grained unlearning in VLMs. Code: https://kodaikawamura.github.io/Domain_Unlearning/.

cs.LG↗