ClickTrack: Towards Real-time Interactive Single Object Tracking
Single object tracking(SOT) relies on precise object bounding box initialization. In this paper, we reconsidered the deficiencies in the current approaches to initializing single object trackers and propose a new paradigm for single object tracking algorithms, ClickTrack, a new paradigm using clicking interaction for real-time scenarios. Moreover, click as an input type inherently lack hierarchical information. To address ambiguity in certain special scenarios, we designed the Guided Click Refiner(GCR), which accepts point and optional textual information as inputs, transforming the point into the bounding box expected by the operator. The bounding box will be used as input of single object trackers. Experiments on LaSOT and GOT-10k benchmarks show that tracker combined with GCR achieves stable performance in real-time interactive scenarios. Furthermore, we explored the integration of GCR into the Segment Anything model(SAM), significantly reducing ambiguity issues when SAM receives point inputs.
Code (0)
등록된 구현이 없습니다.
Tasks
ObjectObject TrackingSemantic SegmentationVideo Object SegmentationVideo Semantic SegmentationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
4DStreamCtrl: Interactive Video Generation with Online 4D Control
Generative video models now synthesize footage nearly indistinguishable from reality. Their promise as interactive tools hinges on fine-grained control of how objects and the camera move over time, yet each existing appr…
Video GenerationCausal InferenceVideo2Game: Real-time, Interactive, Realistic and Browser-Compatible Environment from a Single Video
Creating high-quality and interactive virtual environments, such as games and simulators, often involves complex and costly manual modeling processes. In this paper, we present Video2Game, a novel approach that automatic…
NeRFVideo2Game: Real-time Interactive Realistic and Browser-Compatible Environment from a Single Video
Creating high-quality and interactive virtual environments such as games and simulators often involves complex and costly manual modeling processes. In this paper we present Video2Game a novel approach that automatic…
NeRFSAM-MT: Real-Time Interactive Multi-Target Video Segmentation
Modern Video Object Segmentation (VOS) involves tracking and segmenting user-specified targets. While recent approaches have achieved remarkable performance in single-target scenarios, extending them to multi-target sett…
Video Object SegmentationVideo SegmentationAn Interactive Image-based Modeling System
This paper propose a interactive 3D modeling method and corresponding system based on single or multiple uncalibrated images. The main feature of this method is that, according to the modeling habits of ordinary people, …
Camera Calibration