arXiv subjects

Pengfei Wan

Publications and source records attributed to Pengfei Wan.

At least 19 recordsLinked to original sources

arXiv2026

S3KF: Spherical State-Space Kalman Filtering for Panoramic 3D Multi-Object Tracking

Panoramic multi-object tracking is important for industrial safety monitoring, wide-area robotic perception, and infrastructure-light deployment in large workspaces. In these settings, the sensing system must provide full-surround coverage, metric geometric cues, and stable target association under wide field-of-view distortion and occlusion. Existing image-plane trackers are tightly coupled to the camera projection and become unreliable in panoramic imagery, while conventional Euclidean 3D formulations introduce redundant directional parameters and do not naturally unify angular, scale, and depth estimation. In this paper, we present $\mathbf{S^3KF}$, a panoramic 3D multi-object tracking framework built on a motorized rotating LiDAR and a quad-fisheye camera rig. The key idea is a geometry-consistent state representation on the unit sphere $\mathbb{S}^2$, where object bearing is modeled by a two-degree-of-freedom tangent-plane parameterization and jointly estimated with box scale and depth dynamics. Based on this state, we derive an extended spherical Kalman filtering pipeline that fuses panoramic camera detections with LiDAR depth observations for multimodal tracking. We further establish a map-based ground-truth generation pipeline using wearable localization devices registered to a shared global LiDAR map, enabling quantitative evaluation without motion-capture infrastructure. Experiments on self-collected real-world sequences show decimeter-level planar tracking accuracy, improved identity continuity over a 2D panoramic baseline in dynamic scenes, and real-time onboard operation on a Jetson AGX Orin platform. These results indicate that the proposed framework is a practical solution for panoramic perception and industrial-scale multi-object tracking.The project page can be found at https://kafeiyin00.github.io/S3KF/.

cs.RO↗

arXiv2026

CodeCap-Bench: Can Code Serve as a Good Video Caption?

Video captioning aims to preserve visual information through compact descriptions, predominantly using natural language. However, precisely expressing fine-grained spatial structures and temporal dynamics in natural language often leads to lengthy and potentially ambiguous descriptions. In this work, we explore Code as Video Caption, a paradigm where executable programs serve as a complementary representation of visual content through graphical primitives, layout rules, and functions of time. To examine whether code captions effectively preserve video information, we compare them with natural language captions on downstream question answering and video reconstruction tasks. Our results show that code captions achieve superior reasoning accuracy and reconstruction fidelity. Despite these advantages, realizing the potential of code captioning depends on whether current multimodal large language models (MLLMs) can generate faithful code captions. To systematically assess this capability, we introduce CodeCap-Bench, which comprises 323 videos spanning five categories of structured 2D content, along with fine-grained annotations and a multi-level evaluation framework. Extensive experiments reveal that while leading proprietary MLLMs demonstrate promising code captioning capabilities, they still struggle to faithfully capture complex temporal dynamics, with open-source models lagging substantially behind.

cs.CV↗

arXiv2026

Verus-Insight: Towards Explainable Artifact Detection in AI-Generated Videos

Recent advances in video generation have made synthetic videos increasingly realistic, placing greater demands on reliable AI-generated video detection. Existing artifact-based detectors typically identify generation artifacts as evidence for source prediction, yet they may either fail to recognize subtle artifacts or reach the correct source decision without faithfully identifying the evidence underlying it. In this work, we systematically investigate how to design and train effective artifact detectors and formulate artifact detection as a structured process comprising three complementary capabilities: Inspection Planning, which determines what should be examined; Evidence Verification, which verifies whether the inspected content exhibits genuine generation artifacts; and Evidence Aggregation, which integrates the verified evidence into the final source decision. Building on this formulation, we introduce Verus-Insight, trained with human-anchored full-process supervision and progressive capability training to strengthen both fine-grained artifact detection and video-level detection. We further introduce Verus-Bench, a benchmark with fine-grained human annotations of perceptible artifacts for evaluating artifact detection. Extensive experiments across diverse benchmarks demonstrate that Verus-Insight achieves strong performance in both fine-grained artifact analysis and AI-generated video detection, with substantial gains in recovering human-perceived artifacts. Further analyses reveal that the three capabilities are complementary yet distinct, and that Verus-Insight more reliably converts relevant inspection targets into supported artifact evidence without increasing spurious artifact claims.

cs.CV↗

arXiv2026

RVG-Bench: A Benchmark for Reasoning-Driven Audio-Visual Temporal Grounding

Existing temporal video grounding (TVG) benchmarks primarily focus on localizing target video segments according to descriptive text queries of the visual content. However, real-world queries are more diverse and often require the model to first understand the user's intent before localization, such as finding visual evidence, identifying the cause of an event, or recalling a remembered moment, yet such queries remain largely overlooked. To mitigate this gap, we introduce RVG-Bench, a benchmark for reasoning-driven audio-visual temporal grounding, where every query is constructed such that the target segment can only be identified through reasoning about the query's underlying intent. We carefully construct the benchmark and rigorously verify its data quality to ensure models truly comprehend the audio-visual content. RVG-Bench comprises 1,255 videos and 1,890 query-interval pairs across eight reasoning types. We evaluate 22 models and find that the strongest, Qwen3.8-Omni-Flash, achieves only 49.6 mIoU. Through a series of controlled analyses, including oracle-grounding probes and test-time scaling experiments, we systematically examine where and why current models struggle, providing insights for future model development.

cs.CV↗

arXiv2026

TrajCFG: Token-Wise Classifier-Free Guidance via Denoising Trajectory Feedback

Classifier-free guidance (CFG) typically applies a shared scale to all latent tokens, although their velocity predictions evolve differently during sampling. We introduce Trajectory-Adaptive Classifier-Free Guidance (TrajCFG), a training-free method for flow-matching models that adapts token-wise guidance using historical changes in guided velocity direction. TrajCFG maps this feedback to bounded scales, assigning stronger guidance to larger directional changes and weaker guidance to more consistent directions. It combines adjacent-step and interval-normalized longer-span feedback, reusing standard CFG predictions without additional model evaluations. Controlled diagnostics and ablations examine the roles of directional feedback, token-wise allocation, and history-span fusion. Experiments on text-to-video, image-to-video, text-to-audio-video, text-to-image, and image-to-3D generation show improvements over base samplers across multiple backbones. For text-to-video generation, TrajCFG improves VBench Total scores by 0.30-0.75 points over the strongest evaluated baseline on each of three backbones. These results support historical directional feedback as a broadly applicable signal for fine-grained guidance allocation. The project page is available at https://jyliu-98.github.io/trajcfg-project-page/.

cs.CV↗

arXiv2026

GSHeadRelight: Fast Relightability for 3D Gaussian Head Synthesis

Relighting and novel view synthesis of human portraits are essential in applications such as portrait photography, virtual reality (VR), and augmented reality (AR). Despite recent progress, 3D-aware portrait relighting remains challenging due to the demands for photorealistic rendering, real-time performance, and generalization to unseen subjects. Existing works either rely on supervision from limited and expensive light stage captured data or produce suboptimal results. Moreover, many works are based on generative NeRFs, which suffer from poor 3D consistency and low real-time performance. We resort to recent progress on generative 3D Gaussians and design a lighting model based on a unified neural radiance transfer representation, which responds linearly to incident light. Using only in-the-wild images, our method achieves state-of-the-art relighting results and a significantly faster rendering speed ($\times12$) compared to previous 3D-aware portrait relighting research. Project page: http://www.geometrylearning.com/GSHeadRelight/.

cs.CV↗

arXiv2026

JointLight: Human-Centric Video Relighting with Asynchronous RGB-Irradiance Joint Modeling

Understanding and modeling light transport in human-centric scenes is a longstanding challenge. While recent progress in diffusion models has boosted research in relighting, precise illumination control still lags, primarily due to the scarcity of diverse, high-quality, precisely labeled video relighting datasets. Existing end-to-end relighting approaches model light transport implicitly through neural networks, which substantially complicates controllability. In contrast, some works resort to path-traced irradiance cues as an explicit blueprint for light transport. However, such constructed irradiance cues are prone to artifacts and hinder generative quality when directly used as control signals. With the key observation that noise corruption removes high-frequency artifacts and preserves low-frequency light transport, we introduce an asynchronous joint-modeling paradigm that facilitates fine-grained illumination control. During two-stage training, we force the model to jointly denoise RGB and ground-truth irradiance frames with separate timesteps scheduling. A domain matching strategy is applied to the second stage to bridge the domain gap between irradiance cues and ground-truth irradiance maps. At inference time, we introduce an asynchronous hijacked generation pipeline that facilitates flexible illumination control. For training data, we develop a multi-illumination high-quality synthetic video dataset with paired irradiance maps using a powerful game engine. Trained exclusively on this dataset, our model exhibits robust generalization to real- world scenarios and outperforms prior approaches in both lighting fidelity and generative quality. Project page: http://www.geometrylearning.com/JointLight/

cs.CV↗

arXiv2026

LiveAct: Towards Real-Time Multi-Agent World Models with Heterogeneous Action Interfaces

Interactive video world models predict how an environment evolves under user actions, but most existing systems assume one controlled agent and one task-specific action interface. We study multi-agent video world modeling, where one model generates a joint visual future for multiple controlled agents while interpreting heterogeneous controls ranging from binary game buttons to continuous robot commands. This task presents two complementary challenges. First, every agent-indexed action stream must control its intended visual entity independently. Second, the model must adapt the meaning of each control variable across different tasks, where the same tensor dimension may denote an attack, steering command, or robot command. We introduce LiveAct, centered on a Heterogeneous Action Interface (HAI) whose two components mirror this decomposition. To support independent control, Agent-Aware Spatial Modulation uses language-driven spatial gates and agent-specific attention to bind each action stream to its intended visual agent. To accommodate heterogeneous controls, Schema-Guided Action Adaptation describes each control variable in language and modulates its semantic token with the synchronized numerical value. We further transfer the complete interface to a blockwise causal generator through teacher forcing and rollout-aware distribution matching, enabling efficient streaming generation while retaining agent-specific control and action semantics. We instantiate the formulation on six multi-agent sources spanning games and multi-robot manipulation, and evaluate visual quality, action following, agent attribution, and latency. Across games and robot manipulation, LiveAct reaches 0.898 and 0.942 agent-attribution accuracy without agent locations or identity crops, and the causal generator attains 17.70 FPS at $512\times 448$.

cs.CV↗

arXiv2026

Mise-en-Video: Fine-Grained 3D Control for Multi-Shot Video Generation

Recent advances in text-to-video generation have enabled high-fidelity single-shot synthesis, yet practical storytelling creation requires multi-shot videos with both strong inter-shot consistency and fine-grained controllability. Existing multi-shot video generation approaches achieve strong consistency but still fall short in controllability. In this paper, we propose Mise-en-Video (MeV), a highly controllable multi-shot video generation framework that allows video creators to conduct staging the multi-shot in the 3D space and then generate the corresponding multi-shot video. To achieve this, our framework combines a hierarchical attention mechanism for multi-level narrative consistency with 3D-aware layout guidance for precise, per-shot spatial control. To support training and evaluation, we develop standardized annotation pipelines and construct a 160K-video dataset with hierarchical multi-shot texts and oriented 3D bounding boxes. Extensive experiments demonstrate that MeV consistently outperforms strong baselines in video quality, inter-shot consistency, and subject-level controllability.

cs.CV↗

arXiv2026

PixelMoE: Scaling Pixel-Space Diffusion with LLM-Style Mixture-of-Experts

Pixel-space diffusion models provide an appealing alternative to latent-space diffusion models by directly modeling raw images without relying on a learned tokenizer. However, scaling pixel-space models has predominantly followed dense strategies, tightly coupling model capacity with active computation. In this work, we show that standard LLM-style Mixture-of-Experts (MoE) architectures can be effectively transferred to pixel-space diffusion. Building on this finding, we introduce PixelMoE, a pixel-space generative framework that sparsely scales model capacity with standard LLM-style MoE, without diffusion-specific routing mechanisms or auxiliary routing objectives. Specifically, PixelMoE combines standard Token-Choice (TC) routing with fine-grained routed experts for input-adaptive sparse computation, a shared expert for capturing common representations, and a bias-based load-balancing strategy. To complement sparse capacity scaling, PixelMoE further introduces Pixel-Space Hierarchical Flow Matching, which varies spatial resolution along the denoising trajectory from global structure formation to fine-grained detail refinement. At comparable active parameter counts, PixelMoE consistently outperforms its dense counterparts across model scales, demonstrating effective sparse capacity scaling in pixel space. On class-conditional ImageNet-1K, PixelMoE achieves FID scores of 1.47 at 256x256 and 1.53 at 512x512 without representation alignment or perceptual losses, comparing favorably with leading pixel-space and latent-space generative models. Code and models will be released soon.

cs.CV↗

arXiv2026

PIGEONS: Incremental Long-Video Captioning with Persistent Identity Gallery and Evolving Overarching Narrative Summary

Detailed video captioning has become an essential semantic interface for multi-modal understanding and generation. While existing methods mainly target short videos, extending them to long videos remains challenging, primarily due to limited context windows of current video understanding models. A naive solution is to adopt segment-wise captioning followed by concatenation, yet this paradigm often suffers from inconsistent subject references and fragmented long-range narratives. To mitigate these issues, we introduce PIGEONS, an incremental long-video captioning model built upon a Persistent Identity Gallery and an Evolving Overarching Narrative Summary. Before caption generation, PIGEONS scans the entire video to construct an identity gallery that captures key subjects along with their visual references, thereby providing a stable referential anchor. The model then performs clip-by-clip captioning conditioned on the persistent gallery and an evolving overarching narrative summary, which compactly maintains and updates essential long-range contextual information after each captioning step. To enable these capabilities, we develop a three-stage training strategy that integrates supervised fine-tuning and reinforcement learning. Extensive experiments demonstrate that PIGEONS markedly improves long-video captioning performance, achieving superior accuracy, completeness, and subject referential consistency over existing methods, while maintaining competitive performance on short-video captioning.

cs.CV↗

arXiv2026

EventLens: Event-Structure Reinforcement Learning for Video Understanding

Recent multimodal large language models (MLLMs) achieve strong results on video question answering, yet often fail to recover the temporal structure that makes a video coherent. They may merge neighboring events, miss single-event progression, or infer implausible relations across events. We study structured temporal understanding as the ability to recover event structure from visual evidence, including event boundaries, single-event progression, and multi-event relations. We argue that this gap arises from a mismatch between current post-training objectives and temporal reasoning, where caption-style supervision allows models to rely on language shortcuts rather than visual temporal grounding. To address this, we propose EventLens, a vision-centric reinforcement learning framework that learns video understanding through event-structure recovery. Instead of reconstructing captions, EventLens trains models to recover temporal structure under controlled perturbations, such as altered temporal granularity, reversed progression, and shuffled event order. We instantiate this framework with a three-level temporal ontology and derive three verifiable task families: event segmentation, progression discrimination, and multi-event relation reasoning. These tasks admit deterministic rewards, enabling task-specialized GRPO training without human preference labels or learned reward models. We further introduce multi-teacher on-policy distillation (MT-OPD) to consolidate specialized policies into a unified model. Experiments on Qwen3-VL backbones show consistent improvements on temporal grounding and event-sensitive reasoning benchmarks, while preserving general video-QA performance. Learning curves further demonstrate that EventLens achieves comparable downstream transfer with significantly reduced training cost compared to direct mixed RL optimization.

cs.CV↗

arXiv2026

WorldAlign: Decoupled 4D Reward for World-Consistent Video Generation

Faithful visual world simulation requires generated videos to maintain 4D world consistency, encompassing both static and dynamic consistency. Static consistency requires coherent 3D structure in static environments across viewpoints, while dynamic consistency requires plausible subject motion and consistent appearance over time. Geometry-aware post-training offers a promising way to improve world consistency. However, existing methods often rely on a static-scene assumption. Even those that accommodate dynamic scenes struggle to provide reliable static-consistency feedback, while dynamic consistency is often overlooked or inadequately assessed. To address these limitations, we introduce WorldAlign, a decoupled 4D reward framework that semantically separates static regions and dynamic subjects and provides feedback by aligning each with a world prior suited to its assumptions. For static regions, WorldAlign aligns static geometry with a geometric world prior through semantically guided masked reprojection, enabling more reliable static-consistency evaluation; an auxiliary camera-motion reward discourages nearly static solutions. For dynamic subjects, WorldAlign uses a strong vision-language model (VLM) as a dynamic world prior and constructs a VLM-as-a-judge reward based on sample-specific checklists that assess dynamicity, physical plausibility, shape, and texture consistency. This decoupled design enables more effective online post-training without requiring human preference annotations. Across two pretrained image-to-video generators, Wan2.1 and Wan2.2, WorldAlign jointly improves static and dynamic consistency over existing methods without suppressing overall or subject motion. These results support decoupled world-prior alignment for more faithful visual world simulation. Project page: https://worldalign.github.io/.

cs.CV↗

arXiv2026

DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models

Online reinforcement learning (RL) fine-tunes diffusion models with black-box rewards, but reverse-process methods such as Flow-GRPO are limited in efficiency by poor credit assignment, while forward-process methods such as DiffusionNFT are efficient yet unstable without carefully tuned exponential-moving-average (EMA) off-policy training. We show that NFT-style objectives contain a hidden forward--reverse consistency force that joint positive--negative training cancels only partially, and that EMA stabilizes training through restoration rather than off-policy rollouts. We introduce DiffusionART (Diffusion Reinforcement with Aggregation-First Reward Tilting), which aggregates each prompt's rewards into a shared reward-improving direction for reverse-process states, removing the consistency force by construction, and adds a trust-region-masked second update for drift-adaptive restoration without EMA tuning. It requires no differentiable reward model, remains effective with fewer samples per prompt, and supports pairwise reward models (DiffusionART-Pairwise). Across image and video generation, DiffusionART trains efficiently and stably without per-task tuning. In particular, on HunyuanVideo-13B with equally weighted visual and motion quality rewards, DiffusionART reaches the late-stage mean reward levels of DiffusionNFT and Flow-GRPO-Fast with 12.02 and 11.72 times fewer GPU hours, respectively, while achieving 28.8% and 53.0% higher mean training rewards.

cs.LG↗

arXiv2026

ORCA: The Annealed Spectral Conditioning Optimizer for Faster, Better LLM Training

Modern LLM optimizers such as Muon often produce weight matrices with higher effective rank than Adam, yet further spectral control has delivered only modest gains. We identify a tension behind this result: concentrated spectra can suppress gradient directions in coupled weight matrices and slow optimization, while constraints maintained throughout training can limit task-specific adaptation and raise the attainable loss floor. We introduce ORCA (Orthogonal Regularization, Cooled After), a minimal optimizer intervention that applies strong but temporary soft orthogonality regularization early in training, then removes it. This allows the weights to benefit from a broader spectrum early on and adapt freely afterward. Across LLaMA, Qwen3, and fine-grained mixture-of-experts models ranging from 130M to 8B parameters, ORCA achieves lower final validation loss than Muon. Its loss reduction relative to Muon matches or exceeds Muon's reduction relative to Adam. Ablations support the early-shaping, later-release design. Further, ORCA requires no architectural changes and adds minimal overhead.

cs.LG↗

arXiv2026

Beacon: Knowing When and How to Perform Agentic Visual Reasoning

The fundamental goal of agentic visual reasoning is to improve the success rate of multimodal large language models (MLLMs) on complex tasks. We rethink agentic visual reasoning through two key dimensions of tool use: Mode Adaptiveness and Tool Effect. Mode Adaptiveness characterizes whether an MLLM recognizes when tools are necessary and invokes them accordingly, avoiding unnecessary computational overhead while improving performance on problems requiring tool assistance. Tool Effect characterizes whether tools extend the model's capabilities on problems unsolvable through tool-free reasoning without introducing errors on problems it can already solve. Our analysis quantifies these properties and reveals that existing models exhibit limited Mode Adaptiveness, while tool-use gains on hard examples are largely offset by harm on easy ones. Motivated by these observations, we propose Beacon, a novel agentic visual reasoning model trained with supervised fine-tuning (SFT) and reinforcement learning (RL). Its RL stage combines Necessity-Aware Adaptive Reward and Hint-Guided Capability Expansion. Necessity-Aware Adaptive Reward encourages tool-free solutions when they succeed while preserving full reward for successful tool use when tool-free rollouts fail. Hint-Guided Capability Expansion uses verified, answer-free expert hints to recover learning signals from all-wrong rollout groups, aiming to extend tool-use capability on the hardest problems. Across 13 benchmarks, Beacon achieves the highest average score among the evaluated open-source models and ranks first on 11 benchmarks. On five diagnostic benchmarks, it improves the average tool-available accuracy over its tool-free accuracy by 1.96 points and achieves the largest tool-gain minus tool-harm score (+3.14 points). These results show Beacon's advanced performance, Mode Adaptiveness, and the net benefit of tool use.

cs.CV↗

arXiv2026

Stable Velocity: A Variance Perspective on Flow Matching

While flow matching is elegant, its reliance on single-sample conditional velocities leads to high-variance training targets that destabilize optimization and slow convergence. By explicitly characterizing this variance, we identify 1) a high-variance regime near the prior, where optimization is challenging, and 2) a low-variance regime near the data distribution, where conditional and marginal velocities nearly coincide. Leveraging this insight, we propose Stable Velocity, a unified framework that improves both training and sampling. For training, we introduce Stable Velocity Matching (StableVM), an unbiased variance-reduction objective, along with Variance-Aware Representation Alignment (VA-REPA), which adaptively strengthen auxiliary supervision in the low-variance regime. For inference, we show that dynamics in the low-variance regime admit closed-form simplifications, enabling Stable Velocity Sampling (StableVS), a finetuning-free acceleration. Extensive experiments on ImageNet $256\times256$ and large pretrained text-to-image and text-to-video models, including SD3.5, Flux, Qwen-Image, and Wan2.2, demonstrate consistent improvements in training efficiency and more than $2\times$ faster sampling within the low-variance regime without degrading sample quality. Our code is available at https://github.com/linYDTHU/StableVelocity.

cs.CV↗

arXiv2026

SubjectAnchor: Subject-Aware Memory-to-Video for Multi-Shot Storytelling

We present SubjectAnchor, a Subject-Aware Memory-to-Video paradigm for multi-shot storytelling in which the current shot is generated by conditioning on explicit visual memories extracted from previous shots. The objective is to preserve subject identity and scene consistency across cuts while retaining the controllability of shot-wise prompting. Built on Wan2.2-I2V-A14B, SubjectAnchor contains three key components: subject-related memory construction, subject-aware temporal rotary position encoding, and memory-aware attention partition. For each target shot, the method constructs a compact memory bank by tracing each required subject to its historical appearance and retrieving the most relevant precomputed keyframes. These memory frames are encoded into the model input as explicit visual conditions, while different subjects are assigned to separated negative temporal slots to reduce identity interference. In addition, memory-aware attention partition regulates the interaction between memory tokens and generated content within a shared backbone. This formulation preserves the appearance anchoring of explicit visual memory while remaining compatible with script-driven shot-by-shot generation. Experiments show that SubjectAnchor improves cross-shot identity consistency over representative memory-based and holistic baselines while maintaining competitive visual quality.

cs.CV↗