Recent advances in motion diffusion models have substantially improved the realism of human motion synthesis. However, existing approaches either rely on full-sequence diffusion models with bidirectional generation, which limits temporal causality and real-time applicability, or autoregressive models that suffer from instability and cumulative errors. In this work, we present Causal Motion Diffusion Models (CMDM), a unified framework for autoregressive motion generation based on a causal diffusion transformer that operates in a semantically aligned latent space. CMDM builds upon a Motion-Language-Aligned Causal VAE (MAC-VAE), which encodes motion sequences into temporally causal latent representations. On top of this latent representation, an autoregressive diffusion transformer is trained using causal diffusion forcing to perform temporally ordered denoising across motion frames. To achieve fast inference, we introduce a frame-wise sampling schedule with causal uncertainty, where each subsequent frame is predicted from partially denoised previous frames. The resulting framework supports high-quality text-to-motion generation, streaming synthesis, and long-horizon motion generation at interactive rates. Experiments on HumanML3D and SnapMoGen demonstrate that CMDM outperforms existing diffusion and autoregressive models in both semantic fidelity and temporal smoothness, while substantially reducing inference latency.
While large vision-language models can generate motion graphics animations from text prompts, they regularly fail to include all spatio-temporal properties described in the prompt. We introduce MoVer, a motion verification DSL based on first-order logic that can check spatio-temporal properties of a motion graphics animation. We identify a general set of such properties that people commonly use to describe animations (e.g., the direction and timing of motions, the relative positioning of objects, etc.). We implement these properties as predicates in MoVer and provide an execution engine that can apply a MoVer program to any input SVG-based motion graphics animation. We then demonstrate how MoVer can be used in an LLM-based synthesis and verification pipeline for iteratively refining motion graphics animations. Given a text prompt, our pipeline synthesizes a motion graphics animation and a corresponding MoVer program. Executing the verification program on the animation yields a report of the predicates that failed and the report can be automatically fed back to LLM to iteratively correct the animation. To evaluate our pipeline, we build a synthetic dataset of 5600 text prompts paired with ground truth MoVer verification programs. We find that while our LLM-based pipeline is able to automatically generate a correct motion graphics animation for 58.8% of the test prompts without any iteration, this number raises to 93.6% with up to 50 correction iterations. Our code and dataset are at https://mover-dsl.github.io.
A directed acyclic graph (DAG) can represent a two-dimensional string or picture. We propose recognizing picture languages using DAG automata by encoding 2D inputs into DAGs. An encoding can be input-agnostic (based on input size only) or input-driven (depending on symbols). Three distinct input-agnostic encodings characterize classes of picture languages accepted by returning finite automata, boustrophedon automata, and online tessellation automata. Encoding a string as a simple directed path limits recognition to regular languages. However, input-driven encodings allow DAG automata to recognize some context-sensitive string languages and outperform online tessellation automata in two dimensions.
For safety-critical robotics applications such as autonomous driving, it is important to detect all required objects accurately in real-time. Motion segmentation offers a solution by identifying dynamic objects from the scene in a class-agnostic manner. Recently, various motion segmentation models have been proposed, most of which jointly use subnetworks to estimate Depth, Pose, Optical Flow, and Scene Flow. As a result, the overall computational cost of the model increases, hindering real-time performance. In this paper, we propose a novel cost-volume-based motion feature representation, Channel-wise Motion Features. By extracting depth features of each instance in the feature map and capturing the scene's 3D motion information, it offers enhanced efficiency. The only subnetwork used to build Channel-wise Motion Features is the Pose Network, and no others are required. Our method not only achieves about 4 times the FPS of state-of-the-art models in the KITTI Dataset and Cityscapes of the VCAS-Motion Dataset, but also demonstrates equivalent accuracy while reducing the parameters to about 25$\%$.
The motion of rockets is part of the study devoted to the motion of variable mass systems. Notably those in which the mass leaves permanently the considered system. Rockets are propelled forward by the reaction force produced by the hot exhausted gases ejected from their tales in the rearward direction. Thus their motion should not violate Newton's third principle of the equality of action and reaction forces during the exhaustion process. Nor should it violate Newton's second law of motion judged by inertial observers. However a close examination of the study of the motion of rockets in a major part of physics textbooks, if not all reveals erroneous determination of the expression of the thrust force that pushes the rocket in the forward direction. The false expression of the thrust force entails a bad effect on obtaining the right differential equation that governs the motion of rockets. This trap induced some prominent physics authors to pretend the in applicability of Newton's second law in such particular cases. Not only that, but they also modified Newton's second law in order to fit their purposes of obtaining the right differential equation, historically known under the name Tsiolkovsky rocket equation. The object of this paper is to give the true expression of the thrust force and to write the differential equation of motion of rockets without any necessity of modification of the classical laws. The paper also delves into the expression of the change of velocity of rockets, and proves that their motion is uniformly accelerated in the early stages of liftoff from the ground at condition of constant velocity of expulsion of hot gases from their nozzles.
Eliminating image blur produced by various kinds of motion has been a challenging problem. Dominant approaches rely heavily on model capacity to remove blurring by reconstructing residual from blurry observation in feature space. These practices not only prevent the capture of spatially variable motion in the real world but also ignore the tailored handling of various motions in image space. In this paper, we propose a novel real-world deblurring filtering model called the Motion-adaptive Separable Collaborative (MISC) Filter. In particular, we use a motion estimation network to capture motion information from neighborhoods, thereby adaptively estimating spatially-variant motion flow, mask, kernels, weights, and offsets to obtain the MISC Filter. The MISC Filter first aligns the motion-induced blurring patterns to the motion middle along the predicted flow direction, and then collaboratively filters the aligned image through the predicted kernels, weights, and offsets to generate the output. This design can handle more generalized and complex motion in a spatially differentiated manner. Furthermore, we analyze the relationships between the motion estimation network and the residual reconstruction network. Extensive experiments on four widely used benchmarks demonstrate that our method provides an effective solution for real-world motion blur removal and achieves state-of-the-art performance. Code is available at https://github.com/ChengxuLiu/MISCFilter
Text-driven human motion generation is a multimodal task that synthesizes human motion sequences conditioned on natural language. It requires the model to satisfy textual descriptions under varying conditional inputs, while generating plausible and realistic human actions with high diversity. Existing diffusion model-based approaches have outstanding performance in the diversity and multimodality of generation. However, compared to autoregressive methods that train motion encoders before inference, diffusion methods lack in fitting the distribution of human motion features which leads to an unsatisfactory FID score. One insight is that the diffusion model lack the ability to learn the motion relations among spatio-temporal semantics through contextual reasoning. To solve this issue, in this paper, we proposed Motion Masked Diffusion Model \textbf{(MMDM)}, a novel human motion masked mechanism for diffusion model to explicitly enhance its ability to learn the spatio-temporal relationships from contextual joints among motion sequences. Besides, considering the complexity of human motion data with dynamic temporal characteristics and spatial structure, we designed two mask modeling strategies: \textbf{time frames mask} and \textbf{body parts mask}. During training, MMDM masks certain tokens in the motion embedding space. Then, the diffusion decoder is designed to learn the whole motion sequence from masked embedding in each sampling step, this allows the model to recover a complete sequence from incomplete representations. Experiments on HumanML3D and KIT-ML dataset demonstrate that our mask strategy is effective by balancing motion quality and text-motion consistency.
Mengtian Li, Chengshuo Zhai, Shengxiang Yao
et al.
In the realm of motion generation, the creation of long-duration, high-quality motion sequences remains a significant challenge. This paper presents our groundbreaking work on "Infinite Motion", a novel approach that leverages long text to extended motion generation, effectively bridging the gap between short and long-duration motion synthesis. Our core insight is the strategic extension and reassembly of existing high-quality text-motion datasets, which has led to the creation of a novel benchmark dataset to facilitate the training of models for extended motion sequences. A key innovation of our model is its ability to accept arbitrary lengths of text as input, enabling the generation of motion sequences tailored to specific narratives or scenarios. Furthermore, we incorporate the timestamp design for text which allows precise editing of local segments within the generated sequences, offering unparalleled control and flexibility in motion synthesis. We further demonstrate the versatility and practical utility of "Infinite Motion" through three specific applications: natural language interactive editing, motion sequence editing within long sequences and splicing of independent motion sequences. Each application highlights the adaptability of our approach and broadens the spectrum of possibilities for research and development in motion generation. Through extensive experiments, we demonstrate the superior performance of our model in generating long sequence motions compared to existing methods.Project page: https://shuochengzhai.github.io/Infinite-motion.github.io/
In this article, I engage with Shawn Loht’s argument concerning film viewing as being-in-the-world, developed in his book Phenomenology of Film: A Heideggerian Account of the Film Experience (2017), focusing on the aesthetics of mood with particular attention to boredom. I elaborate on a phenomenological ontology of the film experience and its perceptual “rules” which hinge on aesthetic choices: what kind of world does the film open up for the viewer? Loht’s account of viewing Dasein enables us to deepen phenomenological approaches regarding the relationship between (cinematic) moods and (filmic) understanding, and through his engagement with Mitsein, to expand on film spectatorship conceived as a relation with a shared film world. I look at how Heideggerian boredom can elicit spectatorial experiences and disclosure of meaning, as well as allowing characters’ being in the film world. A broader engagement with cinematic moods is also conducive to exploring limits and potentialities of a possible Heideggerian film-philosophy. To illustrate some of these points, I discuss Ryusuke Hamaguchi’s Drive My Car (2021), wherein boredom is consistently awakened through aesthetic strategies of dead-time, silence, slowness and repetition. In the film, the mood of boredom, but also that of love and grief as world-disclosing moods which produce a radical change in our being-in-the-world, is a way to establish a sense of our embeddedness, orient our understanding and disclose our finitude and solitude.
If film immortalises the ephemeral and presentifies the past, this is especially true of home movies, whose content is not the result of a narrative composition or an invention of fiction, but the product of fragments of reality. These three categories – fiction, documentary, and the home movie – have been analysed by Jean-Pierre Meunier and Vivian Sobchack, with an emphasis on the effect that each film mode can have on the spectator, eliciting a particular emotional and cognitive response. But these authors’ models depend entirely on the spectator’s experience to identify each mode, with the home movie being limited to footage in which we can recognise our own families. Nevertheless, a stranger’s home movie can elicit an act of recognition through an intersubjectivity that arises from the footage. On this basis, this article proposes a reading of the home movie as a reliquary of memory, where content and medium intersect to give rise to the particular experience associated with this film mode.
Differential drive robots that can be modeled as a kinematic unicycle are a standard mobile base platform for many service and logistics robots. Safe and smooth autonomous motion around obstacles is a crucial skill for unicycle robots to perform diverse tasks in complex environments. A classical control approach for unicycle control is feedback linearization using a headway point at a fixed headway distance in front of the unicycle. The unicycle headway control brings the headway point to a desired goal location by embedding a linear headway reference dynamics, which often results in an undesired offset for the actual unicycle position. In this paper, we introduce a new unicycle headway control approach with an adaptive headway distance that overcomes this limitation, i.e., when the headway point reaches the goal the unicycle position is also at the goal. By systematically analyzing the closed-loop unicycle motion under the adaptive headway controller, we design analytical feedback motion prediction methods that bound the closed-loop unicycle position trajectory and so can be effectively used for safety assessment and safe unicycle motion design around obstacles. We present an application of adaptive headway motion control and motion prediction for safe unicycle path following around obstacles in numerical simulations.
In order to build artificial intelligence systems that can perceive and reason with human behavior in the real world, we must first design models that conduct complex spatio-temporal reasoning over motion sequences. Moving towards this goal, we propose the HumanMotionQA task to evaluate complex, multi-step reasoning abilities of models on long-form human motion sequences. We generate a dataset of question-answer pairs that require detecting motor cues in small portions of motion sequences, reasoning temporally about when events occur, and querying specific motion attributes. In addition, we propose NSPose, a neuro-symbolic method for this task that uses symbolic reasoning and a modular design to ground motion through learning motion concepts, attribute neural operators, and temporal relations. We demonstrate the suitability of NSPose for the HumanMotionQA task, outperforming all baseline methods.
This article analyses an important category from Deleuze's philosophy – the notion of intensity – and explores its significance for Deleuze and the ways it can be used to think about poetic cinema. I use the concept of the intensive-image to define a cinematic style that dissipates narrative action in favour of more contemplative and sensory experiences, hence films that are able to turn onscreen reality into purely affective phenomena. The notion of intensity, I argue, does not allow us to reply easily to the question about representation (i.e., “what is the film about?”), mainly because its images are indeterminate and flowing, combined in a multiple manner that excludes any simple reduction to signification. The general argument that follows, illustrated by a careful reading of key poetic texts, is that the focus on the intensive-image inspires readers to reconsider Deleuze's classification of the cinema into two separate periods as well as to understand the ways in which intensity finds expression in films that produce problems, displacements and provocations. In the words of Pasolini, this is the poetic film tradition which must be placed diachronically in relation to the language of the film narrative, a diachronism that, he says, would appear destined to be always more pronounced. In the last section of this article, I look at Luis Buñuel's Un Chien Andalou (1929) as an early case of the poetic image and draw on key theorists and filmmakers who have discussed the cinema's power of affection and intensity.