arXiv subjects

Yunho Kim

Publications and source records attributed to Yunho Kim.

18 recordsLinked to original sources

arXiv2026

3DPhysVideo: Consistency-Guided Flow SDE for Video Generation via 3D Scene Reconstruction and Physical Simulation

Video generative models have made remarkable progress, yet they often yield visual artifacts that violate grounding in physical dynamics. Recent works such as PhysGen3D tackle single image-to-3D physics through mesh reconstruction and Physically-Based Rendering, but challenges remain in modeling fluid dynamics, multi-object interactions and photorealism. This work introduces 3DPhysVideo, a novel training-free pipeline that generates physically realistic videos from a single image. We repurpose an off-the-shelf video model for two stages. First, we use it as a novel view synthesizer to reconstruct complete 360-degree 3D scene geometry by guiding the image-to-video (I2V) flow model with rendered point clouds. Second, after applying physics solvers to this geometry, the physically simulated point cloud is used to guide the same I2V flow model to synthesize final, high-quality videos. Consistency-Guided Flow SDE, which decomposes the predicted velocity of the I2V flow model into denoising and consistency bias, enforces consistency to the conditional inputs, allowing us to effectively repurpose the model for both 3D reconstruction and simulation-guided video generation. In the diverse experiments including multi-objects, and fluid interaction scenes, our method successfully bridges the gap from single-images to physically plausible videos, while remaining efficient to run on a single consumer GPU. It outperforms state-of-the-art baselines on GPT-based scores, VideoPhy benchmark and human evaluation.

cs.CV↗

arXiv2026

Learning-augmented robotic automation for real-world manufacturing

Industrial robots are widely used in manufacturing, yet most manipulation still depends on fixed waypoint scripts that are brittle to environmental changes. Learning-based control offers a more adaptive alternative, but it remains unclear whether such methods, still mostly confined to laboratory demonstrations, can sustain hours of reliable operation, deliver consistent quality, and behave safely around people on a live production line. Here we present Learning-Augmented Robotic Automation, a hybrid system that integrates learned task controllers and a neural 3D safety monitor into conventional industrial workflows. We deployed the system on an electric-motor production line to automate deformable cable insertion and soldering under real manufacturing constraints, a step previously performed manually by human workers. With less than 20 min of real-world data per task, the system operated continuously for 5 h 10 min, producing 108 motors without physical fencing and achieving a 99.4% pass rate on product-level quality-control tests. It maintained near-human takt time while reducing variability in solder-joint quality and cycle time. These results establish a practical pathway for extending industrial automation with learning-based methods.

cs.RO↗

arXiv2026

A space-time dual-pairing summation-by-parts framework for forward and adjoint wave equations

In this paper, we propose the first of its kind space-time dual-pairing summation by parts (DP-SBP) numerical framework for forward and adjoint wave propagation problems. This novel approach enables us to achieve spatial and temporal high order accuracy while naturally introducing dissipation in time. Within this framework, initial and boundary conditions are weakly imposed using the simultaneous approximation term (SAT) technique. Fully discrete energy estimates are derived, ensuring the stability of the resulting numerical scheme. Furthermore, the proposed space-time numerical framework allows us to construct adjoint consistent fully discrete numerical approximations, which can be applied to solve inverse wave propagation problems. We provide numerical experiments in one and two spatial dimensions to verify the theoretical analysis and demonstrate convergence of numerical errors.

math.NA↗

arXiv2025

Learning Fast, Tool aware Collision Avoidance for Collaborative Robots

Ensuring safe and efficient operation of collaborative robots in human environments is challenging, especially in dynamic settings where both obstacle motion and tasks change over time. Current robot controllers typically assume full visibility and fixed tools, which can lead to collisions or overly conservative behavior. In our work, we introduce a tool-aware collision avoidance system that adjusts in real time to different tool sizes and modes of tool-environment interaction. Using a learned perception model, our system filters out robot and tool components from the point cloud, reasons about occluded area, and predicts collision under partial observability. We then use a control policy trained via constrained reinforcement learning to produce smooth avoidance maneuvers in under 10 milliseconds. In simulated and real-world tests, our approach outperforms traditional approaches (APF, MPPI) in dynamic environments, while maintaining sub-millimeter accuracy. Moreover, our system operates with approximately 60% lower computational cost compared to a state-of-the-art GPU-based planner. Our approach provides modular, efficient, and effective collision avoidance for robots operating in dynamic environments. We integrate our method into a collaborative robot application and demonstrate its practical use for safe and responsive operation.

cs.RO↗

arXiv2025

High-speed control and navigation for quadrupedal robots on complex and discrete terrain

High-speed legged navigation in discrete and geometrically complex environments is a challenging task because of the high-degree-of-freedom dynamics and long-horizon, nonconvex nature of the optimization problem. In this work, we propose a hierarchical navigation pipeline for legged robots that can traverse such environments at high speed. The proposed pipeline consists of a planner and tracker module. The planner module finds physically feasible foothold plans by sampling-based optimization with fast sequential filtering using heuristics and a neural network. Subsequently, rollouts are performed in a physics simulation to identify the best foothold plan regarding the engineered cost function and to confirm its physical consistency. This hierarchical planning module is computationally efficient and physically accurate at the same time. The tracker aims to accurately step on the target footholds from the planning module. During the training stage, the foothold target distribution is given by a generative model that is trained competitively with the tracker. This process ensures that the tracker is trained in an environment with the desired difficulty. The resulting tracker can overcome terrains that are more difficult than what the previous methods could manage. We demonstrated our approach using Raibo, our in-house dynamic quadruped robot. The results were dynamic and agile motions: Raibo is capable of running on vertical walls, jumping a 1.3-meter gap, running over stepping stones at 4 meters per second, and autonomously navigating on terrains full of 30{\deg} ramps, stairs, and boxes of various sizes.

cs.RO↗

arXiv2024

Diffusion-Based Offline RL for Improved Decision-Making in Augmented ARC Task

Effective long-term strategies enable AI systems to navigate complex environments by making sequential decisions over extended horizons. Similarly, reinforcement learning (RL) agents optimize decisions across sequences to maximize rewards, even without immediate feedback. To verify that Latent Diffusion-Constrained Q-learning (LDCQ), a prominent diffusion-based offline RL method, demonstrates strong reasoning abilities in multi-step decision-making, we aimed to evaluate its performance on the Abstraction and Reasoning Corpus (ARC). However, applying offline RL methodologies to enhance strategic reasoning in AI for solving tasks in ARC is challenging due to the lack of sufficient experience data in the ARC training set. To address this limitation, we introduce an augmented offline RL dataset for ARC, called Synthesized Offline Learning Data for Abstraction and Reasoning (SOLAR), along with the SOLAR-Generator, which generates diverse trajectory data based on predefined rules. SOLAR enables the application of offline RL methods by offering sufficient experience data. We synthesized SOLAR for a simple task and used it to train an agent with the LDCQ method. Our experiments demonstrate the effectiveness of the offline RL approach on a simple ARC task, showing the agent's ability to make multi-step sequential decisions and correctly identify answer states. These results highlight the potential of the offline RL approach to enhance AI's strategic reasoning capabilities.

cs.AI↗

arXiv2024

DIAR: Diffusion-model-guided Implicit Q-learning with Adaptive Revaluation

We propose a novel offline reinforcement learning (offline RL) approach, introducing the Diffusion-model-guided Implicit Q-learning with Adaptive Revaluation (DIAR) framework. We address two key challenges in offline RL: out-of-distribution samples and long-horizon problems. We leverage diffusion models to learn state-action sequence distributions and incorporate value functions for more balanced and adaptive decision-making. DIAR introduces an Adaptive Revaluation mechanism that dynamically adjusts decision lengths by comparing current and future state values, enabling flexible long-term decision-making. Furthermore, we address Q-value overestimation by combining Q-network learning with a value function guided by a diffusion model. The diffusion model generates diverse latent trajectories, enhancing policy robustness and generalization. As demonstrated in tasks like Maze2D, AntMaze, and Kitchen, DIAR consistently outperforms state-of-the-art algorithms in long-horizon, sparse-reward environments.

cs.LG↗

arXiv2024

A Simple Baseline with Single-encoder for Referring Image Segmentation

Referring image segmentation (RIS) requires dense vision-language interactions between visual pixels and textual words to segment objects based on a given description. However, commonly adapted dual-encoders in RIS, e.g., Swin transformer and BERT (uni-modal encoders) or CLIP (a multi-modal dual-encoder), lack dense multi-modal interactions during pre-training, leading to a gap with a pixel-level RIS task. To bridge this gap, existing RIS methods often rely on multi-modal fusion modules that interact two encoders, but this approach leads to high computational costs. In this paper, we present a novel RIS method with a single-encoder, i.e., BEiT-3, maximizing the potential of shared self-attention across all framework components. This enables seamless interactions of two modalities from input to final prediction, producing granularly aligned multi-modal features. Furthermore, we propose lightweight yet effective decoder modules, a Shared FPN and a Shared Mask Decoder, which contribute to the high efficiency of our model. Our simple baseline with a single encoder achieves outstanding performances on the RIS benchmark datasets while maintaining computational efficiency, compared to the most recent SoTA methods based on dual-encoders.

cs.CV↗

arXiv2024

Not Only Rewards But Also Constraints: Applications on Legged Robot Locomotion

Several earlier studies have shown impressive control performance in complex robotic systems by designing the controller using a neural network and training it with model-free reinforcement learning. However, these outstanding controllers with natural motion style and high task performance are developed through extensive reward engineering, which is a highly laborious and time-consuming process of designing numerous reward terms and determining suitable reward coefficients. In this work, we propose a novel reinforcement learning framework for training neural network controllers for complex robotic systems consisting of both rewards and constraints. To let the engineers appropriately reflect their intent to constraints and handle them with minimal computation overhead, two constraint types and an efficient policy optimization algorithm are suggested. The learning framework is applied to train locomotion controllers for several legged robots with different morphology and physical attributes to traverse challenging terrains. Extensive simulation and real-world experiments demonstrate that performant controllers can be trained with significantly less reward engineering, by tuning only a single reward coefficient. Furthermore, a more straightforward and intuitive engineering process can be utilized, thanks to the interpretability and generalizability of constraints. The summary video is available at https://youtu.be/KAlm3yskhvM.

cs.RO↗

arXiv2024

Learning Semantic Traversability with Egocentric Video and Automated Annotation Strategy

For reliable autonomous robot navigation in urban settings, the robot must have the ability to identify semantically traversable terrains in the image based on the semantic understanding of the scene. This reasoning ability is based on semantic traversability, which is frequently achieved using semantic segmentation models fine-tuned on the testing domain. This fine-tuning process often involves manual data collection with the target robot and annotation by human labelers which is prohibitively expensive and unscalable. In this work, we present an effective methodology for training a semantic traversability estimator using egocentric videos and an automated annotation process. Egocentric videos are collected from a camera mounted on a pedestrian's chest. The dataset for training the semantic traversability estimator is then automatically generated by extracting semantically traversable regions in each video frame using a recent foundation model in image segmentation and its prompting technique. Extensive experiments with videos taken across several countries and cities, covering diverse urban scenarios, demonstrate the high scalability and generalizability of the proposed annotation method. Furthermore, performance analysis and real-world deployment for autonomous robot navigation showcase that the trained semantic traversability estimator is highly accurate, able to handle diverse camera viewpoints, computationally light, and real-world applicable. The summary video is available at https://youtu.be/EUVoH-wA-lA.

cs.RO↗

arXiv2024

Improved weight initialization for deep and narrow feedforward neural network

Appropriate weight initialization settings, along with the ReLU activation function, have become cornerstones of modern deep learning, enabling the training and deployment of highly effective and efficient neural network models across diverse areas of artificial intelligence. The problem of \textquotedblleft dying ReLU," where ReLU neurons become inactive and yield zero output, presents a significant challenge in the training of deep neural networks with ReLU activation function. Theoretical research and various methods have been introduced to address the problem. However, even with these methods and research, training remains challenging for extremely deep and narrow feedforward networks with ReLU activation function. In this paper, we propose a novel weight initialization method to address this issue. We establish several properties of our initial weight matrix and demonstrate how these properties enable the effective propagation of signal vectors. Through a series of experiments and comparisons with existing methods, we demonstrate the effectiveness of the novel initialization method.

cs.LG↗

arXiv2022

DiffFace: Diffusion-based Face Swapping with Facial Guidance

In this paper, we propose a diffusion-based face swapping framework for the first time, called DiffFace, composed of training ID conditional DDPM, sampling with facial guidance, and a target-preserving blending. In specific, in the training process, the ID conditional DDPM is trained to generate face images with the desired identity. In the sampling process, we use the off-the-shelf facial expert models to make the model transfer source identity while preserving target attributes faithfully. During this process, to preserve the background of the target image and obtain the desired face swapping result, we additionally propose a target-preserving blending strategy. It helps our model to keep the attributes of the target face from noise while transferring the source facial identity. In addition, without any re-training, our model can flexibly apply additional facial guidance and adaptively control the ID-attributes trade-off to achieve the desired results. To the best of our knowledge, this is the first approach that applies the diffusion model in face swapping task. Compared with previous GAN-based approaches, by taking advantage of the diffusion model for the face swapping task, DiffFace achieves better benefits such as training stability, high fidelity, diversity of the samples, and controllability. Extensive experiments show that our DiffFace is comparable or superior to the state-of-the-art methods on several standard face swapping benchmarks.

cs.CV↗

arXiv2022

Learning Forward Dynamics Model and Informed Trajectory Sampler for Safe Quadruped Navigation

For autonomous quadruped robot navigation in various complex environments, a typical SOTA system is composed of four main modules -- mapper, global planner, local planner, and command-tracking controller -- in a hierarchical manner. In this paper, we build a robust and safe local planner which is designed to generate a velocity plan to track a coarsely planned path from the global planner. Previous works used waypoint-based methods (e.g. Proportional-Differential control and pure pursuit) which simplify the path tracking problem to local point-goal navigation. However, they suffer from frequent collisions in geometrically complex and narrow environments because of two reasons; the global planner uses a coarse and inaccurate model and the local planner is unable to track the global plan sufficiently well. Currently, deep learning methods are an appealing alternative because they can learn safety and path feasibility from experience more accurately. However, existing deep learning methods are not capable of planning for a long horizon. In this work, we propose a learning-based fully autonomous navigation framework composed of three innovative elements: a learned forward dynamics model (FDM), an online sampling-based model-predictive controller, and an informed trajectory sampler (ITS). Using our framework, a quadruped robot can autonomously navigate in various complex environments without a collision and generate a smoother command plan compared to the baseline method. Furthermore, our method can reactively handle unexpected obstacles on the planned path and avoid them. Project page https://awesomericky.github.io/projects/FDM_ITS_navigation/.

cs.RO↗

arXiv2021

Learning multiple gaits of quadruped robot using hierarchical reinforcement learning

There is a growing interest in learning a velocity command tracking controller of quadruped robot using reinforcement learning due to its robustness and scalability. However, a single policy, trained end-to-end, usually shows a single gait regardless of the command velocity. This could be a suboptimal solution considering the existence of optimal gait according to the velocity for quadruped animals. In this work, we propose a hierarchical controller for quadruped robot that could generate multiple gaits (i.e. pace, trot, bound) while tracking velocity command. Our controller is composed of two policies, each working as a central pattern generator and local feedback controller, and trained with hierarchical reinforcement learning. Experiment results show 1) the existence of optimal gait for specific velocity range 2) the efficiency of our hierarchical controller compared to a controller composed of a single policy, which usually shows a single gait. Codes are publicly available.

cs.RO↗

arXiv2021

Probing Quadratic Gravity with Binary Inspirals

In this paper, we study gravitational waves generated by binary systems within an extension of General Relativity which is described by the addition of quadratic in curvature tensor terms to the Einstein-Hilbert action. Treating quadratic gravity as an effective theory valid in the low energy/curvature regime, we argue that reliable calculations can be performed in the early inspiral phase, and furthermore, no flux of additional massive waves can be detected. We then compute the massive dipole (-1PN) leading corrections to the post-Newtonian (PN) expansion of the standard waveform. By confronting these theoretical calculations with available experimental data, we constrain both unknown parameters of quadratic gravity to be $0 \leq γ\, \lesssim 5.7\cdot 10^{76}$, and $-\fracγ{4} \leq β\, \lesssim - 4.2\cdot 10^{75}$.

gr-qc↗

arXiv2021

Non-Schwarzschild Primordial Black Holes as Dark Matter in Quadratic Gravity

One-loop renormalised quantum effective action for gravity contains quadratic in curvature terms. We have found an approximate analytic black hole solution in quadratic gravity by keeping only the radial spherically symmetric fluctuations and dimensionally reducing the 4-dimensional (4D) theory down to the 2-dimensional (2D) dilaton gravity with a potential. The solution reduces to the Schwarzschild black hole in the limit of Einstein's gravity, but otherwise admits non-negative Arnowitt-Deser-Misner (ADM) and positive quasi-local Misner-Sharp masses that can differ significantly. We then study the thermodynamics of such quantum corrected black holes and compute their lifetime under Hawking evaporation. We note that for some range of parameters, black holes increase in mass while emitting Hawking radiation. This pathological behaviour is related to the negative energy states that are present in quadratic gravity. We also find that the micro-lensing of non-Schwarzschild black holes could significantly deviate from the micro-lensing of their Schwarzschild counterparts. These findings have important ramifications for the phenomenology of primordial black holes (PBHs) as dark matter. In particular, the quoted constraints on PBH dark matter from micro-lensing data can be completely evaded, thus making PBHs in the mass range $\sim 10^{-12} - 10~M_{\odot}$ viable dark matter candidates.

gr-qc↗

arXiv2020

Topologically induced black hole charge and its astrophysical manifestations

Quantum corrected effective action for gravity contains massive spin-2 ghost degrees of freedom and admits a topological term which couples longitudinal vector degrees of freedom of the massive spin-2 to Maxwell's electromagnetism. We argue that in the presence of otherwise unobservable gravimagnetic poles this topological term induces an electric charge on a black hole which can be probed through the associated electric and magnetic fields. In particular, we discuss the electromagnetic follow up from the LIGO-sensitive charged black hole binary coalescence due to the synchrotron radiation from the surrounding plasma and the shadow of super-massive charged black holes.

gr-qc↗

arXiv2017

An unconstrained framework for eigenvalue problems

In this paper, we propose an unconstrained framework for eigenvalue problems in both discrete and continuous settings. We begin our discussion to solve a generalized eigenvalue problem $A{\bf x} = λB{\bf x}$ with two $N\times N$ real symmetric matrices $A, B$ via minimizing a proposed functional whose nonzero critical points ${\bf x}\in\mathbb{R}^N$ solve the eigenvalue problem and whose local minimizers are indeed global minimizers. Inspired by the properties of the proposed functional to be minimized, we provide analysis on convergence of various algorithms either to find critical points or local minimizers. Using the same framework, we will also present an eigenvalue problem for differential operators in the continuous setting. It will be interesting to see that this unconstrained framework is designed to find the smallest eigenvalue through matrix addition and multiplication and that a solution ${\bf x}\in\mathbb{R}^N$ and the matrix $B$ can compute the corresponding eigenvalue $λ$ without using $A$ in the case of $A{\bf x}=λB{\bf x}$. At the end, we will present a few numerical experiments which will confirm our analysis.

math.OC↗