Multiscale Vision Transformers meet Bipartite Matching for efficient single-stage Action Localization

• URL: http://arxiv.org/abs/2312.17686v2
• Date: Thu, 23 May 2024 15:52:11 GMT
• Title: Multiscale Vision Transformers meet Bipartite Matching for efficient single-stage Action Localization
• Authors: Ioanna Ntinou, Enrique Sanchez, Georgios Tzimiropoulos,
• Abstract summary: Action localization is a challenging problem that combines detection and recognition tasks, which are often addressed separately. We show that a single MViTv2-S architecture trained with bipartite matching to perform both tasks surpasses the same MViTv2-S when trained with RoI align on pre-computed bounding boxes.
• Score: 27.472705540825316
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Action Localization is a challenging problem that combines detection and recognition tasks, which are often addressed separately. State-of-the-art methods rely on off-the-shelf bounding box detections pre-computed at high resolution, and propose transformer models that focus on the classification task alone. Such two-stage solutions are prohibitive for real-time deployment. On the other hand, single-stage methods target both tasks by devoting part of the network (generally the backbone) to sharing the majority of the workload, compromising performance for speed. These methods build on adding a DETR head with learnable queries that after cross- and self-attention can be sent to corresponding MLPs for detecting a person's bounding box and action. However, DETR-like architectures are challenging to train and can incur in big complexity. In this paper, we observe that \textbf{a straight bipartite matching loss can be applied to the output tokens of a vision transformer}. This results in a backbone + MLP architecture that can do both tasks without the need of an extra encoder-decoder head and learnable queries. We show that a single MViTv2-S architecture trained with bipartite matching to perform both tasks surpasses the same MViTv2-S when trained with RoI align on pre-computed bounding boxes. With a careful design of token pooling and the proposed training pipeline, our Bipartite-Matching Vision Transformer model, \textbf{BMViT}, achieves +3 mAP on AVA2.2. w.r.t. the two-stage MViTv2-S counterpart. Code is available at \href{https://github.com/IoannaNti/BMViT}{https://github.com/IoannaNti/BMViT}

Related papers

• H$_{2}$OT: Hierarchical Hourglass Tokenizer for Efficient Video Pose Transformers [124.11648300910444]
  We present a hierarchical plug-and-play pruning-and-$-recovering framework, called Hierarchical Hourglass Tokenizer (H$_2$OT)<n>Our method is general-purpose: it can be easily incorporated into common VPT models on both seq2seq and seq2frame pipelines.
  arXiv  Detail & Related papers  (2025-09-08T17:59:59Z)
• Route-and-Execute: Auditable Model-Card Matching and Specialty-Level Deployment [6.7202991099968346]
  We present a framework that uses a single vision-language model (VLM) in two complementary roles.<n>First, the VLM acts as an aware model-card matcher that routes an incoming image to the appropriate specialist model.<n>Second, we fine-tune the VLM on specialty-specific datasets ensuring a single model covers multiple downstream tasks.
  arXiv  Detail & Related papers  (2025-08-22T23:34:37Z)
• Just Functioning as a Hook for Two-Stage Referring Multi-Object Tracking [22.669740476582835]
  Referring Multi-Object Tracking (RMOT) aims to localize target trajectories specified by natural language expressions in videos.<n>We propose JustHook, a novel hook-like framework for two-stage RMOT.
  arXiv  Detail & Related papers  (2025-03-10T16:38:42Z)
• UniMoD: Efficient Unified Multimodal Transformers with Mixture-of-Depths [17.68867710994329]
  UniMoD is a task-aware token pruning method that employs a separate router for each task to determine which tokens should be pruned. We apply our method to Show-o and Emu3, reducing training FLOPs by approximately 15% in Show-o and 40% in Emu3, while maintaining or improving performance on several benchmarks.
  arXiv  Detail & Related papers  (2025-02-10T13:52:52Z)
• Semi-DETR: Semi-Supervised Object Detection with Detection Transformers [105.45018934087076]
  We analyze the DETR-based framework on semi-supervised object detection (SSOD) We present Semi-DETR, the first transformer-based end-to-end semi-supervised object detector. Our method outperforms all state-of-the-art methods by clear margins.
  arXiv  Detail & Related papers  (2023-07-16T16:32:14Z)
• DOAD: Decoupled One Stage Action Detection Network [77.14883592642782]
  Localizing people and recognizing their actions from videos is a challenging task towards high-level video understanding. Existing methods are mostly two-stage based, with one stage for person bounding box generation and the other stage for action recognition. We present a decoupled one-stage network dubbed DOAD, to improve the efficiency for-temporal action detection.
  arXiv  Detail & Related papers  (2023-04-01T08:06:43Z)
• HyRSM++: Hybrid Relation Guided Temporal Set Matching for Few-shot Action Recognition [51.2715005161475]
  We propose a novel Hybrid Relation guided temporal Set Matching approach for few-shot action recognition. The core idea of HyRSM++ is to integrate all videos within the task to learn discriminative representations. We show that our method achieves state-of-the-art performance under various few-shot settings.
  arXiv  Detail & Related papers  (2023-01-09T13:32:50Z)
• MulT: An End-to-End Multitask Learning Transformer [66.52419626048115]
  We propose an end-to-end Multitask Learning Transformer framework, named MulT, to simultaneously learn multiple high-level vision tasks. Our framework encodes the input image into a shared representation and makes predictions for each vision task using task-specific transformer-based decoder heads.
  arXiv  Detail & Related papers  (2022-05-17T13:03:18Z)
• BatchFormerV2: Exploring Sample Relationships for Dense Representation Learning [88.82371069668147]
  BatchFormerV2 is a more general batch Transformer module, which enables exploring sample relationships for dense representation learning. BatchFormerV2 consistently improves current DETR-based detection methods by over 1.3%.
  arXiv  Detail & Related papers  (2022-04-04T05:53:42Z)
• CvT-ASSD: Convolutional vision-Transformer Based Attentive Single Shot MultiBox Detector [15.656374849760734]
  We present a novel object detection architecture, named Convolutional vision Transformer Based Attentive Single Shot MultiBox Detector (CvT-ASSD) Our model CvT-ASSD can leads to good system efficiency and performance while being pretrained on large-scale detection datasets such as PASCAL VOC and MS COCO.
  arXiv  Detail & Related papers  (2021-10-24T06:45:33Z)
• A Co-Interactive Transformer for Joint Slot Filling and Intent Detection [61.109486326954205]
  Intent detection and slot filling are two main tasks for building a spoken language understanding (SLU) system. Previous studies either model the two tasks separately or only consider the single information flow from intent to slot. We propose a Co-Interactive Transformer to consider the cross-impact between the two tasks simultaneously.
  arXiv  Detail & Related papers  (2020-10-08T10:16:52Z)

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.