ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1512.02413
40
14
v1v2v3 (latest)

Tracking Objects with Higher Order Interactions using Delayed Column Generation

8 December 2015
Shaofei Wang
Steffen Wolf
Charless C. Fowlkes
ArXiv (abs)PDFHTML
Abstract

We introduce a new approach to tracking a large number of objects. Specifically we consider tracking in the context of higher order Markov interactions which can not be modeled using network flow techniques as currently developed. Our approach relies on delayed column generation and is inspired by the corresponding approach to the cutting stock problem. Columns can be generated exactly using dynamic programming.

View on arXiv
Comments on this paper