mlsanityData to 14 Oct 2026

Open-World Object Counting in Videos

arXiv 2506.15368 ↗First posted 18 Jun 2025cs.CV cs.AIcited by 11

Not on the map. Why?

Abstract (arXiv)

We introduce a new task of open-world object counting in videos: given a text description, or an image example, that specifies the target object, the objective is to enumerate all the unique instances of the target objects in the video. This task is especially challenging in crowded scenes with occlusions and objects of similar appearance, where avoiding double counting and identifying reappearances is crucial. To this end, we make the following contributions: we introduce a model, CountVid, for this task. It leverages an image-based counting model, and a promptable video segmentation and tracking model, to enable automated open-world object counting across video frames. To evaluate its performance, we introduce VideoCount, a new dataset for this novel task built from the TAO and MOT20 tracking datasets, as well as from videos of penguins and metal alloy crystallization captured by x-rays. Using this dataset, we demonstrate that CountVid provides accurate object counts, and significantly outperforms strong baselines. The VideoCount dataset, the CountVid model, and all the code are available at https://www.robots.ox.ac.uk/~vgg/research/countvid/.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 40 of its 60 references that are in the map. First posted on arXiv 18 Jun 2025.

Builds on

40 of 60 references.

References · 40

  1. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  2. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,332
  3. 2024SAM 2: Segment Anything in Images and Videoscited by 2,863
  4. 2021ByteTrack: Multi-Object Tracking by Associating Every Detection Boxcited by 823
  5. 2022Detecting Twenty-thousand Classes using Image-level Supervisioncited by 614
  6. 2021TrackFormer: Multi-Object Tracking with Transformerscited by 518
  7. 2020MOT20: A benchmark for multi object tracking in crowded scenescited by 357
  8. 2023Scaling Open-Vocabulary Object Detectioncited by 343
  9. 2021Towards Open World Object Detectioncited by 297
  10. 2020Detection and Tracking Meet Drones Challengecited by 263
  11. 2019The VIA Annotation Software for Images, Audio and Videocited by 167
  12. 2017Drone-based Object Counting by Spatially Regularized Regional Proposal Networkcited by 156
28 more
  1. 2021Learning To Count Everythingcited by 150
  2. 2020TAO: A Large-Scale Benchmark for Tracking Any Objectcited by 147
  3. 2024SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memorycited by 94
  4. 2018Class-Agnostic Countingcited by 85
  5. 2023CLIP-Count: Towards Text-Guided Zero-Shot Object Countingcited by 83
  6. 2022Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Countingcited by 80
  7. 2022CounTR: Transformer-based Generalised Visual Countingcited by 79
  8. 2022A Low-Shot Object Counting Network With Iterative Prototype Adaptationcited by 65
  9. 2022Few-shot Object Counting with Similarity-Aware Feature Enhancementcited by 62
  10. 2024CountGD: Multi-Modal Open-World Countingcited by 59
  11. 2022Few-shot Object Counting and Detectioncited by 55
  12. 2024Matching Anything by Segmenting Anythingcited by 50
  13. 2024DAVE -- A Detect-and-Verify Paradigm for Low-Shot Countingcited by 49
  14. 2023VLCounter: Text-aware Visual Representation for Zero-Shot Object Countingcited by 45
  15. 2021Tracking Pedestrian Heads in Dense Crowdcited by 42
  16. 2023Open-world Text-specified Object Countingcited by 41
  17. 2023Point, Segment and Count: A Generalized Framework for Object Countingcited by 39
  18. 2021Detection, Tracking, and Counting Meets Drones in Crowds: A Benchmarkcited by 31
  19. 2019Locality-constrained Spatial Transformer Network for Video Crowd Countingcited by 30
  20. 2022The Caltech Fish Counting Dataset: A Benchmark for Multiple-Object Tracking and Countingcited by 24
  21. 2024A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentationcited by 21
  22. 2022DR.VIC: Decomposition and Reasoning for Video Individual Countingcited by 18
  23. 2023Video OWL-ViT: Temporally-consistent open-world localization in videocited by 13
  24. 2025OVTR: End-to-End Open-Vocabulary Multiple Object Tracking with Transformercited by 11
  25. 2024OVR: A Dataset for Open Vocabulary Temporal Repetition Counting in Videoscited by 10
  26. 2024Every Shot Counts: Using Exemplars for Repetition Counting in Videoscited by 9
  27. 2024SLAck: Semantic, Location, and Appearance Aware Open-Vocabulary Trackingcited by 7
  28. 2024VOVTrack: Exploring the Potentiality in Videos for Open-Vocabulary Object Trackingcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 10.

11 papers.

  1. 2025CountGD++: Generalized Prompting for Open-World Countingcited by 8
  2. 2026MotionCFG: Boosting Motion Dynamics via Stochastic Concept Perturbationcited by 1
  3. 2026Automated Counting of Stacked Objects in Industrial Inspectioncited by 1
  4. 2026The MixCount Dataset: Bridging the Data Gap for Open-Vocabulary Object Countingcited by 1
  5. 2026AdaCount: Training-Free Similarity-Guided Spatial and Feature Adaptation for Zero-Shot Object Countingcited by 1
  6. 2026A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understandingcited by 0
  7. 2026Referring Multiple Regions with Large Multimodal Models via Contextual Latent Steeringcited by 0
  8. 2026Count Anythingcited by 0
  9. 2026Depth-Guided Video Object Counting in Crowded Scenescited by 0
  10. 2026Object Counting Across Modalities: Taxonomies, Benchmarks, Applications, and Open Challengescited by 0
  11. 2026UGO: Unified Architecture for General Multi-Object Tracking by Segmentationcited by 0