About

Abstract: Conventional multi-object tracking (MOT) systems are predominantly designed for pedestrian tracking and often exhibit limited generalization to other object categories. This paper presents a generalized tracking framework capable of handling multiple object types, with a particular emphasis on vehicle tracking in complex traffic scenes. The proposed method incorporates two key components: (1) an occlusion-aware re- identification mechanism that enhances identity preservation for heavily occluded objects, and (2) a road-structure-aware tracklet refinement strategy that utilizes semantic scene priors—such as lane directions, crosswalks, and road boundaries—to improve trajectory continuity and accuracy. In addition, we introduce a new benchmark dataset comprising diverse vehicle classes with frame-level tracking annotations, specifically curated to support evaluation of vehicle-focused tracking methods. Extensive experimental results demonstrate that the proposed approach achieves robust performance on both the newly introduced dataset and several public benchmarks, highlighting its effectiveness in general-purpose object tracking. While our framework is designed for generalized multi-class tracking, it also achieves strong per- formance on conventional benchmarks, with HOTA scores of 66.4 on MOT17 and 65.7 on MOT20 test sets. github.com/Hamidreza- Hashempoor/FastTracker, huggingface.co/datasets/Hamidreza- Hashemp/FastTracker-Benchmark.

Abstract: We propose a framework for structured temporal inference in nonlinear state-space models (SSMs) with hybrid latent dynamics that mix discrete and continuous variables. Our method follows a two-stage inference: continuous states are estimated via Kalman inspired updates, while discrete variables are sampled by a neural model conditioned on these states, avoiding explicit Markov assumptions. To handle instabilities arising from recurrent dynamics, we introduce stabilization approach, and train all components jointly using surrogate gradient estimators that support REINFORCE-style updates. This design achieves SOTA results across synthetic and real-world datasets, in state estimation, regime detection, and imputation under noise and partial observability.

Abstract: We introduce FeatureSORT, a simple yet effective online multiple object tracker that reinforces the baselines with a redesigned detector and additional feature cues, while keeping computational complexity low. In contrast to conventional detectors that only provide bounding boxes, our designed detector architecture is extended to output multiple appearance attributes, including clothing color, clothing style, and motion direction, alongside the bounding boxes. These feature cues, together with a ReID network, form complementary embeddings that substantially improve association accuracy. The rationale behind selecting and combining these attributes is thoroughly examined in extensive ablation studies. Furthermore, we incorporate stronger post-processing strategies, such as global linking and Gaussian Smoothing Process interpolation, to handle missing associations and detections. During online tracking, we define a measurement-to-track distance function that jointly considers IoU, direction, color, style, and ReID similarity. This design enables FeatureSORT to maintain consistent identities through longer occlusions while reducing identity switches. Extensive experiments on standard MOT benchmarks demonstrate that FeatureSORT achieves state-of-the-art (SOTA) online performance, with MOTA scores of 79.7 on MOT16, 80.6 on MOT17, 77.9 on MOT20, and 92.2 on DanceTrack, underscoring the effectiveness of feature-enriched detection in advancing multi-object tracking. Our Github repository includes code implementation.

Abstract: A multi-camera multi-target tracking framework that assigns globally consistent identities using trajectory and appearance cues, with glance-based initialization and progressive global association.

Abstract: Proposes a deep learning-based data-assisted channel estimation and detection framework for MIMO-OFDM systems, with denoising and correctness classification blocks that improve estimation MSE and detection quality.

Abstract: Introduces GIN, an approximate Bayesian inference algorithm for nonlinear state-space models that disentangles object and dynamic latent representations for state estimation and missing data imputation.

Abstract: Presents an online MOT tracker that combines multiple appearance cues and ReID with stronger detection and post-processing, improving identity consistency and occlusion robustness.

Abstract: Proposes GaussianNet, a Gaussian stochastic network for selecting reliable detected symbols in data-assisted channel estimation, reducing MSE and BER versus conventional methods.

Abstract: Provides a benchmark deep learning-from-demonstration pipeline and dataset for needle insertion tasks with da Vinci Research Kit, linking visual observations to robot control trajectories.

Abstract: Introduces a synchronized multi-camera dataset and baseline deep RLfD architectures for robotic needle insertion through deformable objects, supporting visual-to-action learning.