MetaDesigner introduces a transformative framework for artistic typography synthesis, powered by Large Language Models (LLMs) and grounded in a user-centric design paradigm. Its foundation is a multi-agent system comprising the Pipeline, Glyph, and Texture agents, which collectively orchestrate the creation of customizable WordArt, ranging from semantic enhancements to intricate textural elements. A central feedback mechanism leverages insights from both multimodal models and user evaluations, enabling iterative refinement of design parameters. Through this iterative process, MetaDesigner dynamically adjusts hyperparameters to align with user-defined stylistic and thematic preferences, consistently delivering WordArt that excels in visual quality and contextual resonance. Empirical evaluations underscore the system's versatility and effectiveness across diverse WordArt applications, yielding outputs that are both aesthetically compelling and context-sensitive.

Text logo design heavily relies on the creativity and expertise of professional designers, in which arranging element layouts is one of the most important procedures. However, this specific task has received limited attention, often overshadowed by broader layout generation tasks such as document or poster design. In this paper, we propose a Vision-Language Model (VLM)-based framework that generates content-aware text logo layouts by integrating multi-modal inputs with user-defined constraints, enabling more flexible and robust layout generation for real-world applications. We introduce two model techniques that reduce the computational cost for processing multiple glyph images simultaneously, without compromising performance. To support instruction tuning of our model, we construct two extensive text logo datasets that are five times larger than existing public datasets. In addition to geometric annotations (e.g., text masks and character recognition), our datasets include detailed layout descriptions in natural language, enabling the model to reason more effectively in handling complex designs and custom user inputs. Experimental results demonstrate the effectiveness of our proposed framework and datasets, outperforming existing methods on various benchmarks that assess geometric aesthetics and human preferences.

The advancement of autonomous driving systems hinges on the ability to achieve low-latency and high-accuracy perception. To address this critical need, this paper introduces Dynamic Routering Network (DyRoNet), a low-rank enhanced dynamic routing framework designed for streaming perception in autonomous driving systems. DyRoNet integrates a suite of pre-trained branch networks, each meticulously fine-tuned to function under distinct environmental conditions. At its core, the framework offers a speed router module, developed to assess and route input data to the most suitable branch for processing. This approach not only addresses the inherent limitations of conventional models in adapting to diverse driving conditions but also ensures the balance between performance and efficiency. Extensive experimental evaluations demonstrating the adaptability of DyRoNet to diverse branch selection strategies, resulting in significant performance enhancements across different scenarios. This work not only establishes a new benchmark for streaming perception but also provides valuable engineering insights for future work.11Project: https://tastevision.github.io/DyRoNet/

The speed-precision tradeoff is a critical problem in visual object tracking, as it typically requires low latency and is deployed on resource-constrained platforms. Existing solutions for efficient tracking primarily focus on lightweight backbones or modules, which, however, come at a sacrifice in precision. In this article, inspired by dynamic network routing, we propose DyTrack, a dynamic transformer framework for efficient tracking. Real-world tracking scenarios exhibit varying levels of complexity. We argue that a simple network is sufficient for easy video frames, while more computational resources should be assigned to difficult ones. DyTrack automatically learns to configure proper reasoning routes for different inputs, thereby improving the utilization of the available computational budget and achieving higher performance at the same running speed. We formulate instance-specific tracking as a sequential decision problem and incorporate terminating branches to intermediate layers of the model. Furthermore, we propose a feature recycling mechanism to maximize computational efficiency by reusing the outputs of predecessors. Additionally, a target-aware self-distillation strategy is designed to enhance the discriminating capabilities of early-stage predictions by mimicking the representation patterns of the deep model. Extensive experiments demonstrate that DyTrack achieves promising speed-precision tradeoffs with only a single model. For instance, DyTrack obtains 64.9% area under the curve (AUC) on LaSOT with a speed of 256 fps.

For the skeleton-based gesture recognition, graph convolutional networks (GCNs) have achieved remarkable performance since the human skeleton is a natural graph. However, the biological structure might not be the crucial one for motion analysis. Also, spatial differential information like joint distance and angle between bones may be overlooked during the graph convolution. In this article, we focus on obtaining meaningful joint groups and extracting their discriminative features by the path signature (PS) theory. Firstly, to characterize the constraints and dependencies of various joints, we propose three types of paths, i.e., spatial, temporal, and learnable path. Especially, a learnable path generation mechanism can group joints together that are not directly connected or far away, according to their kinematic characteristic. Secondly, to obtain informative and compact features, a deep integration of PS with few parameters are introduced. All the computational process is packed into two modules, i.e., spatial-temporal path signature module (ST-PSM) and learnable path signature module (L-PSM) for the convenience of utilization. They are plug-and-play modules available for any neural network like CNNs and GCNs to enhance the feature extraction ability. Extensive experiments have conducted on three mainstream datasets (ChaLearn 2013, ChaLearn 2016, and AUTSL). We achieved the state-of-the-art results with simpler framework and much smaller model size. By inserting our two modules into the several GCN-based networks, we can observe clear improvements demonstrating the great effectiveness of our proposed method.

Existing Visual Object Tracking (VOT) only takes the target area in the first frame as a template. This causes tracking to inevitably fail in fast-changing and crowded scenes, as it cannot account for changes in object appearance between frames. To this end, we revamped the tracking framework with Progressive Context Encoding Transformer Tracker (ProContEXT), which coherently exploits spatial and temporal contexts to predict object motion trajectories. Specifically, ProContEXT leverages a context-aware self-attention module to encode the spatial and temporal context, refining and updating the multi-scale static and dynamic templates to progressively perform accurately tracking. It explores the complementary between spatial and temporal context, raising a new pathway to multi-context modeling for transformer-based trackers. In addition, ProContEXT revised the token pruning technique to reduce computational complexity. Extensive experiments on popular benchmark datasets such as GOT-10k and TrackingNet demonstrate that the proposed ProContEXT achieves state-of-the-art performance1.

Streaming perception is a fundamental task in autonomous driving that requires a careful balance between the latency and accuracy of the autopilot system. However, current methods for streaming perception are limited as they rely only on the current and adjacent two frames to learn movement patterns, which restricts their ability to model complex scenes, often leading to poor detection results. To address this limitation, we propose LongShortNet, a novel dual-path network that captures long-term temporal motion and integrates it with short-term spatial semantics for real-time perception. Our proposed LongShortNet is notable as it is the first work to extend long-term temporal modeling to streaming perception, enabling spatiotemporal feature fusion. We evaluate LongShortNet on the challenging Argoverse-HD dataset and demonstrate that it outperforms existing state-of-the-art methods with almost no additional computational cost.1

In the realm of autonomous driving, real-time perception or streaming perception remains under-explored. This research introduces DAMO-StreamNet, a novel framework that merges the cutting-edge elements of the YOLO series with a detailed examination of spatial and temporal perception techniques. DAMO-StreamNet's main inventions include: (1) a robust neck structure employing deformable convolution, bolstering receptive field and feature alignment capabilities; (2) a dual-branch structure synthesizing short-path semantic features and long-path temporal features, enhancing the accuracy of motion state prediction; (3) logits-level distillation facilitating efficient optimization, which aligns the logits of teacher and student networks in semantic space; and (4) a real-time prediction mechanism that updates the features of support frames with the current frame, providing smooth streaming perception during inference. Our testing shows that DAMO-StreamNet surpasses current state-of-the-art methodologies, achieving 37.8% (normal size (600, 960)) and 43.3% (large size (1200, 1920)) sAP without requiring additional data. This study not only establishes a new standard for real-time perception but also offers valuable insights for future research. The source code is at https://github.com/zhiqic/DAMO-StreamNet.

In the realm of facial analysis, accurate landmark detection is crucial for various applications, ranging from face recognition and expression analysis to animation. Conventional heatmap or coordinate regression-based techniques, however, often face challenges in terms of computational burden and quantization errors. To address these issues, we present the KeyPoint Positioning System (KeyPosS) - a groundbreaking facial landmark detection framework that stands out from existing methods. The framework utilizes a fully convolutional network to predict a distance map, which computes the distance between a Point of Interest (POI) and multiple anchor points. These anchor points are ingeniously harnessed to triangulate the POI's position through the True-range Multilateration algorithm. Notably, the plug-and-play nature of KeyPosS enables seamless integration into any decoding stage, ensuring a versatile and adaptable solution. We conducted a thorough evaluation of KeyPosS's performance by benchmarking it against state-of-the-art models on four different datasets. The results show that KeyPosS substantially outperforms leading methods in low-resolution settings while requiring a minimal time overhead.1 The code is available at https://github.com/zhiqic/KeyPosS.

Studies have shown that there is a tight connection between cognition skills and brain morphology during infancy. Nonetheless, it is still a great challenge to predict individual cognitive scores using their brain morphological features, considering issues like the excessive feature dimension, small sample size and missing data. Due to the limited data, a compact but expressive feature set is desirable as it can reduce the dimension and avoid the potential overfitting issue. Therefore, we pioneer the path signature method to further explore the essential hidden dynamic patterns of longitudinal cortical features. To form a hierarchical and more informative temporal representation, in this work, a novel cortical feature based path signature neural network (CF-PSNet) is proposed with stacked differentiable temporal path signature layers for prediction of individual cognitive scores. By introducing the existence embedding in path generation, we can improve the robustness against the missing data. Benefiting from the global temporal receptive field of CF-PSNet, characteristics consisted in the existing data can be fully leveraged. Further, as there is no need for the whole brain to work for a certain cognitive ability, a top K selection module is used to select the most influential brain regions, decreasing the model size and the risk of overfitting. Extensive experiments are conducted on an in-house longitudinal infant dataset within 9 time points. By comparing with several recent algorithms, we illustrate the state-of-the-art performance of our CF-PSNet (i.e., root mean square error of 0.027 with the time latency of 518 milliseconds for each sample).

The skeleton based gesture recognition is gaining more popularity due to its wide possible applications. The key issues are how to extract discriminative features and how to design the classification model. In this paper, we first leverage a robust feature descriptor, path signature (PS), and propose three PS features to explicitly represent the spatial and temporal motion characteristics, i.e., spatial PS (S PS), temporal PS (T PS) and temporal spatial PS (T S PS). Considering the significance of fine hand movements in the gesture, we propose an ”attention on hand” (AOH) principle to define joint pairs for the S PS and select single joint for the T PS. In addition, the dyadic method is employed to extract the T PS and T S PS features that encode global and local temporal dynamics in the motion. Secondly, without the recurrent strategy, the classification model still faces challenges on temporal variation among different sequences. We propose a new temporal transformer module (TTM) that can match the sequence key frames by learning the temporal shifting parameter for each input. This is a learning-based module that can be included into standard neural network architecture. Finally, we design a multi-stream fully connected layer based network to treat spatial and temporal features separately and fused them together for the final result. We have tested our method on three benchmark gesture datasets, i.e., ChaLearn 2016, ChaLearn 2013 and MSRC-12. Experimental results demonstrate that we achieve the state-of-the-art performance on skeleton-based gesture recognition with high computational efficiency.

Skeleton based gesture recognition has gained more attention due to its wide application and large-scale databases availability. Recent methods designed for skeleton sequence data mainly pay attention to network architecture but ignore an essential characteristic of skeleton sequences that the temporal dimensionality of skeleton sequences is usually higher than its spatial dimensionality. Directly applying CNNs designed for image classification to skeleton-based data can not capture this unique property. Considering this fact, we propose the rectangular convolution and pooling to skeleton sequence data. Temporal features are crucial for gesture action recognition. Further, we introduce path signature features (PSF) to represent temporal variation characteristics of each joint. Moreover, there only exist a few minor distinctions between some gestures. To classify them more accurately, we add two sub-networks to extract discriminative features from two hands respectively. We evaluate our method on three major benchmark gesture datasets, i.e., ChaLearn 2013, ChaLearn 2016 and MSRC-12, and reach the state-of-the-art performance.