Author name cluster

Kun Tian

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

2 author rows

NeurIPS Conference 2025 Conference Paper

StreamForest: Efficient Online Video Understanding with Persistent Event Memory

Xiangyu Zeng
Kefan Qiu
Qingyu Zhang
Xinhao Li
Jing Wang
Jiaxin Li
Ziang Yan
Kun Tian

Multimodal Large Language Models (MLLMs) have recently achieved remarkable progress in video understanding. However, their effectiveness in real-time streaming scenarios remains limited due to storage constraints of historical visual features and insufficient real-time spatiotemporal reasoning. To address these challenges, we propose StreamForest, a novel architecture specifically designed for streaming video understanding. Central to StreamForest is the Persistent Event Memory Forest, a memory mechanism that adaptively organizes video frames into multiple event-level tree structures. This process is guided by penalty functions based on temporal distance, content similarity, and merge frequency, enabling efficient long-term memory retention under limited computational resources. To enhance real-time perception, we introduce a Fine-grained Spatiotemporal Window, which captures detailed short-term visual cues to improve current scene perception. Additionally, we present OnlineIT, an instruction-tuning dataset tailored for streaming video tasks. OnlineIT significantly boosts MLLM performance in both real-time perception and future prediction. To evaluate generalization in practical applications, we introduce ODV-Bench, a new benchmark focused on real-time streaming video understanding in autonomous driving scenarios. Experimental results demonstrate that StreamForest achieves the state-of-the-art performance, with accuracies of 77. 3% on StreamingBench, 60. 5% on OVBench, and 55. 6% on OVO-Bench. In particular, even under extreme visual token compression (limited to 1024 tokens), the model retains 96. 8% of its average accuracy in eight benchmarks relative to the default setting. These results underscore the robustness, efficiency, and generalizability of StreamForest for streaming video understanding.

PDF Details

EAAI Journal 2024 Journal Article

Deep learning model for the deformation prediction of concrete dams under multistep and multifeature inputs based on an improved autoformer

Kun Tian
Jie Yang
Lin Cheng

The long-term prediction of deformation in concrete dams is a critical requirement for maintaining their structural integrity over time in practical management scenarios. While most existing models predominantly address short-term and medium-term predictions, there remains a significant challenge in establishing reliable correlations within complex long-term temporal patterns and varying environmental factors. Consequently, there is limited research on multistep prediction. To solve these problems, an Autoformer-based model for multistep and multifeature dam deformation prediction is proposed. In the proposed model, the sequence decomposition of dam monitoring data is used as the basic construction module inside the depth model. Then, Deep Automatic Correlation (Deep-AutoCorrelation) mechanism is employed to obtain the long-term dependence between dam deformation and environmental load. The encoder-decoder structure integrates time series decomposition and the Deep-AutoCorrelation to predict the multistep deformation of the dam. In the multistep deformation prediction of the dam, the improved Autoformer model demonstrates state-of-the-art performance, resulting in a 48% increase in prediction accuracy across five monitoring point datasets when compared to six benchmark models. Experimental results indicate that this method effectively addresses the challenge of predicting long-term deformation of dams subjected to complex environmental loads. It demonstrates robust capabilities in extracting long-term temporal features and avoids the increase of computational complexity caused by deeper time extraction.

Details DOI

AAAI Conference 2024 Conference Paper

DiffBEV: Conditional Diffusion Model for Bird’s Eye View Perception

Jiayu Zou
Kun Tian
Zheng Zhu
Yun Ye
Xingang Wang

BEV perception is of great importance in the field of autonomous driving, serving as the cornerstone of planning, controlling, and motion prediction. The quality of the BEV feature highly affects the performance of BEV perception. However, taking the noises in camera parameters and LiDAR scans into consideration, we usually obtain BEV representation with harmful noises. Diffusion models naturally have the ability to denoise noisy samples to the ideal data, which motivates us to utilize the diffusion model to get a better BEV representation. In this work, we propose an end-to-end framework, named DiffBEV, to exploit the potential of diffusion model to generate a more comprehensive BEV representation. To the best of our knowledge, we are the first to apply diffusion model to BEV perception. In practice, we design three types of conditions to guide the training of the diffusion model which denoises the coarse samples and refines the semantic feature in a progressive way. What's more, a cross-attention module is leveraged to fuse the context of BEV feature and the semantic content of conditional diffusion model. DiffBEV achieves a 25.9% mIoU on the nuScenes dataset, which is 6.2% higher than the best-performing existing approach. Quantitative and qualitative results on multiple benchmarks demonstrate the effectiveness of DiffBEV in BEV semantic segmentation and 3D object detection tasks.

PDF Details DOI

ICRA Conference 2023 Conference Paper

Efficient and Hybrid Decoder for Local Map Construction in Bird'-Eye-View

Kun Tian
Yun Ye
Zheng Zhu
Peng Li
Guan Huang 0003

High-definition maps are crucial perception elements for autonomous robot navigation systems, which can provide accurate scene layout and environment information for downstream motion prediction and planning control tasks. Traditional methods based on manual annotation or SLAM algorithms require massive labor efforts and time costs, which hinders the deployment of practical applications. Online construction of local maps from on-board cameras offers an alternative solution. Aiming at the problems of unsatisfying precision and redundant computation of HDMapNet, we propose an efficient and hybrid decoder (EHD) that consists of a CNN-based segmentation (Seg) head and a query-based lane detection head (QLD). Specifically, the Seg head outputs pixel-level semantic maps, and QLD predicts instance mask for each lane object through learnable query embeddings. The designed decoding method eliminates the cumulative error caused by inaccurate semantic maps and does not require additional clustering algorithm for post-processing. Through combining with a variety of bird's-eye-view (BEV) encoders, the effectiveness and efficiency of our EHD is demonstrated by extensive experiments. For segmentation task, the mIoU scores of semantic map can be improved by 1. 3%∼2. 9%. Additionally, the accuracy of lane detection is also significantly increased (more than 10. 2% mAP) under all evaluation criteria. Since our method discards redundant post-processing, the inference speed is up to 22. 71 FPS, which is 32 times faster than HDMapNet.

Details

AAAI Conference 2020 Conference Paper

End-to-End Thorough Body Perception for Person Search

Kun Tian
Houjing Huang
Yun Ye
Shiyu Li
Jinbin Lin
Guan Huang

In this paper, we propose an improved end-to-end multibranch person search network to jointly optimize person detection, re-identiﬁcation, instance segmentation, and keypoint detection. First, we build a better and faster base model to extract non-highly correlated feature expression; Second, a foreground feature enhance module is used to alleviate undesirable background noise in person feature maps; Third, we design an algorithm to learn the part-aligned representation for person search. Extensive experiments with ablation analysis show the effectiveness of our proposed end-to-end multi-task model, and we demonstrate its superiority over the state-of-the-art methods on two benchmark datasets including CUHK-SYSU and PRW.

PDF Details