PuTR-CouT: Counting-by-Tracking in Camera-Trap Image Sequences

• URL: http://arxiv.org/abs/2609.05038v1
• Date: Fri, 04 Sep 2026 11:57:22 GMT
• Title: PuTR-CouT: Counting-by-Tracking in Camera-Trap Image Sequences
• Abstract summary: PuTR-CouT is a counting-by-tracking framework built on a transformer-based learned association mechanism for sequence-level animal counting in camera trap images.<n>Our framework delivers competitive counting results compared to the improved MaxBoxCount, with the added capability of multi-species predictions and track-level verification.
• Score: 1.139527693693141
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Species identification in camera trap images has been widely studied, but key ecological modeling tasks such as species abundance or density estimation also require counting individual animals. However, the lack of counting labels in most datasets and low frame rates (typically ~1 frame per second) make sequence-level tracking and count estimation particularly challenging. In this work, we present PuTR-CouT, a counting-by-tracking framework built on a transformer-based learned association mechanism for sequence-level animal counting in camera trap images. To address the scarcity of annotated tracking data, we generate synthetic training data by exploiting structural priors, such as static backgrounds and short temporal bursts, to heuristically create pseudo-tracking labels in a weakly supervised manner. The resulting tracker associates detections across frames, using these tracks to estimate per-species counts. We also refine the MaxBoxCount heuristic used by the top solutions of the iWildCam 2021 challenge as a strong baseline, setting the highest score reported to date. When evaluated on the iWildCam 2021 benchmark, our framework PuTR-CouT delivers competitive counting results compared to the improved MaxBoxCount, with the added capability of multi-species predictions and track-level verification.

Related papers

• BMCTrack-d: Pig re-identification and tracking via back marks in challenging camera settings [0.2770822269241974]
  BMCTrack-d is presented, a novel tracking-by-detection approach that leverages unique back marks.<n>The method first predicts the detected pigs' identities using a neural network-based back mark.<n> BMCTrack-d outperforms two strong baselines on a demanding test set.
  arXiv  Detail & Related papers  (2026-09-03T07:20:03Z)
• Counting Animals in Camera-Traps Image Sequences without Count Labels: Winning Solution to the iWildCam 2021 Challenge [1.139527693693141]
  We describe the winning solution to the iWildCam 2021 Challenge, which introduced a benchmark for counting animals at the sequence level.<n>Our approach, MaxBoxCount, combines a strong species classification pipeline with a simple yet effective counting based on MegaDetector detections.
  arXiv  Detail & Related papers  (2026-09-03T00:29:15Z)
• Promptable Animal Pose Tracking Across Species [6.856674886457795]
  Animal pose estimation and tracking is important for wildlife monitoring and conservation research.<n>We show that vision foundation models trained on large datasets can be used effectively to track animal pose with limited labelled data.<n>We propose two models, one unsupervised and the other supervised, to track user-selected keypoints in videos.
  arXiv  Detail & Related papers  (2026-08-05T16:04:20Z)
• Self-Supervised Animal Identification for Long Videos [0.8233028449337972]
  We introduce a highly efficient, self-supervised method that reframes animal identification as a global clustering task.<n>Our framework matches or surpasses supervised baselines trained on over 1,000 labeled frames.<n>This work enables practical, high-accuracy animal identification on consumer-grade hardware.
  arXiv  Detail & Related papers  (2026-01-14T17:53:59Z)
• BlinkTrack: Feature Tracking over 80 FPS via Events and Images [50.98675227695814]
  We propose a novel framework, BlinkTrack, which integrates event data with grayscale images for high-frequency feature tracking.<n>Our method extends the traditional Kalman filter into a learning-based framework, utilizing differentiable Kalman filters in both event and image branches.<n> Experimental results indicate that BlinkTrack significantly outperforms existing methods, exceeding 80 FPS with multi-modality data and 100 FPS with preprocessed event data.
  arXiv  Detail & Related papers  (2024-09-26T15:54:18Z)
• APTv2: Benchmarking Animal Pose Estimation and Tracking with a Large-scale Dataset and Beyond [27.50166679588048]
  APTv2 is the pioneering large-scale benchmark for animal pose estimation and tracking. It comprises 2,749 video clips filtered and collected from 30 distinct animal species. We provide high-quality keypoint and tracking annotations for a total of 84,611 animal instances.
  arXiv  Detail & Related papers  (2023-12-25T04:49:49Z)
• Tracking by Associating Clips [110.08925274049409]
  In this paper, we investigate an alternative by treating object association as clip-wise matching. Our new perspective views a single long video sequence as multiple short clips, and then the tracking is performed both within and between the clips. The benefits of this new approach are two folds. First, our method is robust to tracking error accumulation or propagation, as the video chunking allows bypassing the interrupted frames. Second, the multiple frame information is aggregated during the clip-wise matching, resulting in a more accurate long-range track association than the current frame-wise matching.
  arXiv  Detail & Related papers  (2022-12-20T10:33:17Z)
• APT-36K: A Large-scale Benchmark for Animal Pose Estimation and Tracking [77.87449881852062]
  APT-36K is the first large-scale benchmark for animal pose estimation and tracking. It consists of 2,400 video clips collected and filtered from 30 animal species with 15 frames for each video, resulting in 36,000 frames in total. We benchmark several representative models on the following three tracks: (1) supervised animal pose estimation on a single frame under intra- and inter-domain transfer learning settings, (2) inter-species domain generalization test for unseen animals, and (3) animal pose estimation with animal tracking.
  arXiv  Detail & Related papers  (2022-06-12T07:18:36Z)
• Visible-Thermal UAV Tracking: A Large-Scale Benchmark and New Baseline [80.13652104204691]
  In this paper, we construct a large-scale benchmark with high diversity for visible-thermal UAV tracking (VTUAV) We provide a coarse-to-fine attribute annotation, where frame-level attributes are provided to exploit the potential of challenge-specific trackers. In addition, we design a new RGB-T baseline, named Hierarchical Multi-modal Fusion Tracker (HMFT), which fuses RGB-T data in various levels.
  arXiv  Detail & Related papers  (2022-04-08T15:22:33Z)
• Unsupervised Learning of Accurate Siamese Tracking [68.58171095173056]
  We present a novel unsupervised tracking framework, in which we can learn temporal correspondence both on the classification branch and regression branch. Our tracker outperforms preceding unsupervised methods by a substantial margin, performing on par with supervised methods on large-scale datasets such as TrackingNet and LaSOT.
  arXiv  Detail & Related papers  (2022-04-04T13:39:43Z)
• TAO: A Large-Scale Benchmark for Tracking Any Object [95.87310116010185]
  Tracking Any Object dataset consists of 2,907 high resolution videos, captured in diverse environments, which are half a minute long on average. We ask annotators to label objects that move at any point in the video, and give names to them post factum. Our vocabulary is both significantly larger and qualitatively different from existing tracking datasets.
  arXiv  Detail & Related papers  (2020-05-20T21:07:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.