Yifu Zhang

Yifu Zhang, Xinggang Wang, Xiaoqing Ye et al.

Multi-object tracking (MOT) aims at estimating bounding boxes and identities of objects across video frames. Detection boxes serve as the basis of both 2D and 3D MOT. The inevitable changing of detection scores leads to object missing after tracking. We propose a hierarchical data association strategy to mine the true objects in low-score detection boxes, which alleviates the problems of object missing and fragmented trajectories. The simple and generic data association strategy shows effectiveness under both 2D and 3D settings. In 3D scenarios, it is much easier for the tracker to predict object velocities in the world coordinate. We propose a complementary motion prediction strategy that incorporates the detected velocities with a Kalman filter to address the problem of abrupt motion and short-term disappearing. ByteTrackV2 leads the nuScenes 3D MOT leaderboard in both camera (56.4% AMOTA) and LiDAR (70.1% AMOTA) modalities. Furthermore, it is nonparametric and can be integrated with various detectors, making it appealing in real applications. The source code is released at https://github.com/ifzhang/ByteTrack-V2.

Yifu Zhang, Chunyu Wang, Xinggang Wang et al.

Multi-object tracking in videos requires to solve a fundamental problem of one-to-one assignment between objects in adjacent frames. Most methods address the problem by first discarding impossible pairs whose feature distances are larger than a threshold, followed by linking objects using Hungarian algorithm to minimize the overall distance. However, we find that the distribution of the distances computed from Re-ID features may vary significantly for different videos. So there isn't a single optimal threshold which allows us to safely discard impossible pairs. To address the problem, we present an efficient approach to compute a marginal probability for each pair of objects in real time. The marginal probability can be regarded as a normalized distance which is significantly more stable than the original feature distance. As a result, we can use a single threshold for all videos. The approach is general and can be applied to the existing trackers to obtain about one point improvement in terms of IDF1 metric. It achieves competitive results on MOT17 and MOT20 benchmarks. In addition, the computed probability is more interpretable which facilitates subsequent post-processing operations.

Jiahui Fu, Chengyuan Lin, Yuichi Taguchi et al.

The ability to process environment maps across multiple sessions is critical for robots operating over extended periods of time. Specifically, it is desirable for autonomous agents to detect changes amongst maps of different sessions so as to gain a conflict-free understanding of the current environment. In this paper, we look into the problem of change detection based on a novel map representation, dubbed Plane Signed Distance Fields (PlaneSDF), where dense maps are represented as a collection of planes and their associated geometric components in SDF volumes. Given point clouds of the source and target scenes, we propose a three-step PlaneSDF-based change detection approach: (1) PlaneSDF volumes are instantiated within each scene and registered across scenes using plane poses; 2D height maps and object maps are extracted per volume via height projection and connected component analysis. (2) Height maps are compared and intersected with the object map to produce a 2D change location mask for changed object candidates in the source scene. (3) 3D geometric validation is performed using SDF-derived features per object candidate for change mask refinement. We evaluate our approach on both synthetic and real-world datasets and demonstrate its effectiveness via the task of changed object detection. Supplementary video: https://youtu.be/oh-MQPWTwZI

Ying Guo, Qijun Gan, Yifu Zhang et al.

Video generation is rapidly evolving towards unified audio-video generation. In this paper, we present ALIVE, a generation model that adapts a pretrained Text-to-Video (T2V) model to Sora-style audio-video generation and animation. In particular, the model unlocks the Text-to-Video&Audio (T2VA) and Reference-to-Video&Audio (animation) capabilities compared to the T2V foundation models. To support the audio-visual synchronization and reference animation, we augment the popular MMDiT architecture with a joint audio-video branch which includes TA-CrossAttn for temporally-aligned cross-modal fusion and UniTemp-RoPE for precise audio-visual alignment. Meanwhile, a comprehensive data pipeline consisting of audio-video captioning, quality control, etc., is carefully designed to collect high-quality finetuning data. Additionally, we introduce a new benchmark to perform a comprehensive model test and comparison. After continue pretraining and finetuning on million-level high-quality data, ALIVE demonstrates outstanding performance, consistently outperforming open-source models and matching or surpassing state-of-the-art commercial solutions. With detailed recipes and benchmarks, we hope ALIVE helps the community develop audio-video generation models more efficiently. Official page: https://github.com/FoundationVision/Alive.

Jincheng Lu, Xipeng Yang, Jin Ye et al.

Multi-Camera Multi-Target Tracking (MCMT) is a computer vision technique that involves tracking multiple targets simultaneously across multiple cameras. MCMT in urban traffic visual analysis faces great challenges due to the complex and dynamic nature of urban traffic scenes, where multiple cameras with different views and perspectives are often used to cover a large city-scale area. Targets in urban traffic scenes often undergo occlusion, illumination changes, and perspective changes, making it difficult to associate targets across different cameras accurately. To overcome these challenges, we propose a novel systematic MCMT framework, called CityTrack. Specifically, we present a Location-Aware SCMT tracker which integrates various advanced techniques to improve its effectiveness in the MCMT task and propose a novel Box-Grained Matching (BGM) method for the ICA module to solve the aforementioned problems. We evaluated our approach on the public test set of the CityFlowV2 dataset and achieved an IDF1 of 84.91%, ranking 1st in the 2022 AI CITY CHALLENGE. Our experimental results demonstrate the effectiveness of our approach in overcoming the challenges posed by urban traffic scenes.

Xiaofan Li, Yifu Zhang, Xiaoqing Ye

With the increasing popularity of autonomous driving based on the powerful and unified bird's-eye-view (BEV) representation, a demand for high-quality and large-scale multi-view video data with accurate annotation is urgently required. However, such large-scale multi-view data is hard to obtain due to expensive collection and annotation costs. To alleviate the problem, we propose a spatial-temporal consistent diffusion framework DrivingDiffusion, to generate realistic multi-view videos controlled by 3D layout. There are three challenges when synthesizing multi-view videos given a 3D layout: How to keep 1) cross-view consistency and 2) cross-frame consistency? 3) How to guarantee the quality of the generated instances? Our DrivingDiffusion solves the problem by cascading the multi-view single-frame image generation step, the single-view video generation step shared by multiple cameras, and post-processing that can handle long video generation. In the multi-view model, the consistency of multi-view images is ensured by information exchange between adjacent cameras. In the temporal model, we mainly query the information that needs attention in subsequent frame generation from the multi-view images of the first frame. We also introduce the local prompt to effectively improve the quality of generated instances. In post-processing, we further enhance the cross-view consistency of subsequent frames and extend the video length by employing temporal sliding window algorithm. Without any extra cost, our model can generate large-scale realistic multi-camera driving videos in complex urban scenes, fueling the downstream driving tasks. The code will be made publicly available.

Yifu Zhang, Zuozhu Liu, Yang Feng et al.

Accurate representation of tooth position is extremely important in treatment. 3D dental image segmentation is a widely used method, however labelled 3D dental datasets are a scarce resource, leading to the problem of small samples that this task faces in many cases. To this end, we address this problem with a pretrained SAM and propose a novel 3D-U-SAM network for 3D dental image segmentation. Specifically, in order to solve the problem of using 2D pre-trained weights on 3D datasets, we adopted a convolution approximation method; in order to retain more details, we designed skip connections to fuse features at all levels with reference to U-Net. The effectiveness of the proposed method is demonstrated in ablation experiments, comparison experiments, and sample size experiments.

Yuyang Zhang, Yifu Zhang, Xuehai Zhou et al.

While empirical scaling laws for LLM reasoning are well-documented, the theoretical mechanisms governing out-of-distribution (OOD) generalization remain elusive. We formalize reasoning via optimal transport, projecting discrete trajectories into a continuous metric space to quantify domain shifts using the Wasserstein-1 distance. Invoking Kantorovich duality, we bound OOD generalization via architectural Lipschitz continuity and functional approximation limits. This exposes two primary constraints. First, position-dependent attention (e.g., Absolute Positional Encoding) fails to preserve shift invariance, yielding an $Ω(1)$ Lipschitz constant and expected risk, whereas shift-invariant mechanisms (e.g., Rotary Embeddings) preserve equivariance and bound the error. Second, by mapping sequential backtracking to a Dyck-$k$ language, we establish a strict circuit depth lower bound for $\text{TC}^0$ Transformers. Scaling physical layer depth is necessary to avert representation collapse -- a constraint that scaling representation width cannot bypass due to irreducible approximation bounds in Barron spaces. Evaluations across 54 Transformer configurations on combinatorial search corroborate these bounds, demonstrating that generalization risk degrades monotonically with the Wasserstein domain shift.

Yifu Zhang, Hao Yang, Yuqi Zhang et al.

We present Waver, a high-performance foundation model for unified image and video generation. Waver can directly generate videos with durations ranging from 5 to 10 seconds at a native resolution of 720p, which are subsequently upscaled to 1080p. The model simultaneously supports text-to-video (T2V), image-to-video (I2V), and text-to-image (T2I) generation within a single, integrated framework. We introduce a Hybrid Stream DiT architecture to enhance modality alignment and accelerate training convergence. To ensure training data quality, we establish a comprehensive data curation pipeline and manually annotate and train an MLLM-based video quality model to filter for the highest-quality samples. Furthermore, we provide detailed training and inference recipes to facilitate the generation of high-quality videos. Building on these contributions, Waver excels at capturing complex motion, achieving superior motion amplitude and temporal consistency in video synthesis. Notably, it ranks among the Top 3 on both the T2V and I2V leaderboards at Artificial Analysis (data as of 2025-07-30 10:00 GMT+8), consistently outperforming existing open-source models and matching or surpassing state-of-the-art commercial solutions. We hope this technical report will help the community more efficiently train high-quality video generation models and accelerate progress in video generation technologies. Official page: https://github.com/FoundationVision/Waver.

Haoran Shen, Yifu Zhang, Wenxuan Wang et al.

Recent works have shown that the computational efficiency of 3D medical image (e.g. CT and MRI) segmentation can be impressively improved by dynamic inference based on slice-wise complexity. As a pioneering work, a dynamic architecture network for medical volumetric segmentation (i.e. Med-DANet) has achieved a favorable accuracy and efficiency trade-off by dynamically selecting a suitable 2D candidate model from the pre-defined model bank for different slices. However, the issues of incomplete data analysis, high training costs, and the two-stage pipeline in Med-DANet require further improvement. To this end, this paper further explores a unified formulation of the dynamic inference framework from the perspective of both the data itself and the model structure. For each slice of the input volume, our proposed method dynamically selects an important foreground region for segmentation based on the policy generated by our Decision Network and Crop Position Network. Besides, we propose to insert a stage-wise quantization selector to the employed segmentation model (e.g. U-Net) for dynamic architecture adapting. Extensive experiments on BraTS 2019 and 2020 show that our method achieves comparable or better performance than previous state-of-the-art methods with much less model complexity. Compared with previous methods Med-DANet and TransBTS with dynamic and static architecture respectively, our framework improves the model efficiency by up to nearly 4.1 and 17.3 times with comparable segmentation results on BraTS 2019.

Jiang-Tian Zhai, Ze Feng, Jinhao Du et al.

Modern autonomous driving systems are typically divided into three main tasks: perception, prediction, and planning. The planning task involves predicting the trajectory of the ego vehicle based on inputs from both internal intention and the external environment, and manipulating the vehicle accordingly. Most existing works evaluate their performance on the nuScenes dataset using the L2 error and collision rate between the predicted trajectories and the ground truth. In this paper, we reevaluate these existing evaluation metrics and explore whether they accurately measure the superiority of different methods. Specifically, we design an MLP-based method that takes raw sensor data (e.g., past trajectory, velocity, etc.) as input and directly outputs the future trajectory of the ego vehicle, without using any perception or prediction information such as camera images or LiDAR. Our simple method achieves similar end-to-end planning performance on the nuScenes dataset with other perception-based methods, reducing the average L2 error by about 20%. Meanwhile, the perception-based methods have an advantage in terms of collision rate. We further conduct in-depth analysis and provide new insights into the factors that are critical for the success of the planning task on nuScenes dataset. Our observation also indicates that we need to rethink the current open-loop evaluation scheme of end-to-end autonomous driving in nuScenes. Codes are available at https://github.com/E2E-AD/AD-MLP.

Yifu Zhang, Peize Sun, Yi Jiang et al.

Multi-object tracking (MOT) aims at estimating bounding boxes and identities of objects in videos. Most methods obtain identities by associating detection boxes whose scores are higher than a threshold. The objects with low detection scores, e.g. occluded objects, are simply thrown away, which brings non-negligible true object missing and fragmented trajectories. To solve this problem, we present a simple, effective and generic association method, tracking by associating almost every detection box instead of only the high score ones. For the low score detection boxes, we utilize their similarities with tracklets to recover true objects and filter out the background detections. When applied to 9 different state-of-the-art trackers, our method achieves consistent improvement on IDF1 score ranging from 1 to 10 points. To put forwards the state-of-the-art performance of MOT, we design a simple and strong tracker, named ByteTrack. For the first time, we achieve 80.3 MOTA, 77.3 IDF1 and 63.1 HOTA on the test set of MOT17 with 30 FPS running speed on a single V100 GPU. ByteTrack also achieves state-of-the-art performance on MOT20, HiEve and BDD100K tracking benchmarks. The source code, pre-trained models with deploy versions and tutorials of applying to other trackers are released at https://github.com/ifzhang/ByteTrack.

Yifu Zhang, Chunyu Wang, Xinggang Wang et al.

Multi-object tracking (MOT) is an important problem in computer vision which has a wide range of applications. Formulating MOT as multi-task learning of object detection and re-ID in a single network is appealing since it allows joint optimization of the two tasks and enjoys high computation efficiency. However, we find that the two tasks tend to compete with each other which need to be carefully addressed. In particular, previous works usually treat re-ID as a secondary task whose accuracy is heavily affected by the primary detection task. As a result, the network is biased to the primary detection task which is not fair to the re-ID task. To solve the problem, we present a simple yet effective approach termed as FairMOT based on the anchor-free object detection architecture CenterNet. Note that it is not a naive combination of CenterNet and re-ID. Instead, we present a bunch of detailed designs which are critical to achieve good tracking results by thorough empirical studies. The resulting approach achieves high accuracy for both detection and tracking. The approach outperforms the state-of-the-art methods by a large margin on several public datasets. The source code and pre-trained models are released at https://github.com/ifzhang/FairMOT.

Yifu Zhang, Hongru Li, Tao Yang et al.

Lesion segmentation of ultrasound medical images based on deep learning techniques is a widely used method for diagnosing diseases. Although there is a large amount of ultrasound image data in medical centers and other places, labeled ultrasound datasets are a scarce resource, and it is likely that no datasets are available for new tissues/organs. Transfer learning provides the possibility to solve this problem, but there are too many features in natural images that are not related to the target domain. As a source domain, redundant features that are not conducive to the task will be extracted. Migration between ultrasound images can avoid this problem, but there are few types of public datasets, and it is difficult to find sufficiently similar source domains. Compared with natural images, ultrasound images have less information, and there are fewer transferable features between different ultrasound images, which may cause negative transfer. To this end, a multi-source adversarial transfer learning network for ultrasound image segmentation is proposed. Specifically, to address the lack of annotations, the idea of adversarial transfer learning is used to adaptively extract common features between a certain pair of source and target domains, which provides the possibility to utilize unlabeled ultrasound data. To alleviate the lack of knowledge in a single source domain, multi-source transfer learning is adopted to fuse knowledge from multiple source domains. In order to ensure the effectiveness of the fusion and maximize the use of precious data, a multi-source domain independent strategy is also proposed to improve the estimation of the target domain data distribution, which further increases the learning ability of the multi-source adversarial migration learning network in multiple domains.

Yifu Zhang, Hongru Li, Shimeng Shi et al.

Transfer learning leverages knowledge from other domains and has been successful in many applications. Transfer learning methods rely on the overall similarity of the source and target domains. However, in some cases, it is impossible to provide an overall similar source domain, and only some source domains with similar local features can be provided. Can transfer learning be achieved? In this regard, we propose a multi-source adversarial transfer learning method based on local feature similarity to the source domain to handle transfer scenarios where the source and target domains have only local similarities. This method extracts transferable local features between a single source domain and the target domain through a sub-network. Specifically, the feature extractor of the sub-network is induced by the domain discriminator to learn transferable knowledge between the source domain and the target domain. The extracted features are then weighted by an attention module to suppress non-transferable local features while enhancing transferable local features. In order to ensure that the data from the target domain in different sub-networks in the same batch is exactly the same, we designed a multi-source domain independent strategy to provide the possibility for later local feature fusion to complete the key features required. In order to verify the effectiveness of the method, we made the dataset "Local Carvana Image Masking Dataset". Applying the proposed method to the image segmentation task of the proposed dataset achieves better transfer performance than other multi-source transfer learning methods. It is shown that the designed transfer learning method is feasible for transfer scenarios where the source and target domains have only local similarities.

Yifu Zhang, Chunyu Wang, Xinggang Wang et al.

We present VoxelTrack for multi-person 3D pose estimation and tracking from a few cameras which are separated by wide baselines. It employs a multi-branch network to jointly estimate 3D poses and re-identification (Re-ID) features for all people in the environment. In contrast to previous efforts which require to establish cross-view correspondence based on noisy 2D pose estimates, it directly estimates and tracks 3D poses from a 3D voxel-based representation constructed from multi-view images. We first discretize the 3D space by regular voxels and compute a feature vector for each voxel by averaging the body joint heatmaps that are inversely projected from all views. We estimate 3D poses from the voxel representation by predicting whether each voxel contains a particular body joint. Similarly, a Re-ID feature is computed for each voxel which is used to track the estimated 3D poses over time. The main advantage of the approach is that it avoids making any hard decisions based on individual images. The approach can robustly estimate and track 3D poses even when people are severely occluded in some cameras. It outperforms the state-of-the-art methods by a large margin on three public datasets including Shelf, Campus and CMU Panoptic.

Yifu Zhang, Jorge Cortes

This paper proposes a control strategy for power systems with a two-layer structure that achieves global stabilization and, at the same time, delimits the transient frequencies of targeted buses to a desired safe interval. The first layer is a model predictive control that, in a receding horizon fashion, optimally allocates the power resources while softly respecting transient frequency constraints. As the first layer control requires solving an optimization problem online, it only periodically samples the system state and updates its action. The second layer control, however, is implemented in real time, assisting the first layer to achieve frequency invariance and attractivity requirements.We show that the controllers designed at both layers are Lipschitz in the state. Furthermore, through network partition, they can be implemented in a distributed fashion, only requiring system information from neighboring partitions. Simulations on the IEEE 39-bus network illustrate our results.

Yifu Zhang, Jorge Cortes

This paper proposes a distributed strategy regulated on a subset of individual buses in a power network described by the swing equations to achieve transient frequency control while preserving asymptotic stability. Transient frequency control refers to the ability to maintain the transient frequency of each bus of interest in a given safe region, provided it is initially in it, and ii) if it is initially not, then drive the frequency to converge to this region within a finite time, with a guaranteed convergence rate. Building on Lyapunov stability and set invariance theory, we formulate the stability and the transient frequency requirements as two separate constraints for the control input. Our design synthesizes a controller that satisfies both constraints simultaneously. The controller is distributed and Lipschitz, guaranteeing the existence and uniqueness of the trajectories of the closed-loop system. We further bound its magnitude and demonstrate its robustness against measurement inaccuracies. Simulations on the IEEE 39-bus power network illustrate our results.

Yifu Zhang, Jorge Cortes

This paper proposes a centralized and a distributed sub-optimal control strategy to maintain in safe regions the real-time transient frequencies of a given collection of buses, and simultaneously preserve asymptotic stability of the entire network. In a receding horizon fashion, the centralized control input is obtained by iteratively solving an open-loop optimization aiming to minimize the aggregate control effort over controllers regulated on individual buses with transient frequency and stability constraints. Due to the non-convexity of the optimization, we propose a convexification technique by identifying a reference control input trajectory. We then extend the centralized control to a distributed scheme, where each subcontroller can only access the state information within a local region. Simulations on a IEEE-39 network illustrate our results.

Yifu Zhang, Jorge Cortes

Modern power networks face increasing challenges in controlling their transient frequency behavior at acceptable levels due to low inertia and highly-dynamic units. This paper presents a distributed control strategy regulated on a subset of buses in a power network to maintain their transient frequencies in safe regions while preserving asymptotic stability of the overall system. Building on Lyapunov stability and set invariance theory, we formulate the transient frequency requirement and the asymptotic stability requirement as two separate constraints for the control input. Hereby, for each bus of interest, we synthesize a controller satisfying both constraints simultaneously. The controller is distributed and Lipschitz, guaranteeing the existence and uniqueness of the trajectories of the closed-loop system. Simulations on the IEEE 39-bus power network illustrate the results.