Trajectory Inference

Follow this topic Ask about this topic

Trajectory inference aims to recover how people, vehicles, robots, or biological populations move and change over time when observations are sparse, incomplete, or unpaired. For population dynamics, common approaches include optimal transport, Schrödinger bridges, flow matching, and neural differential equations; geometric and lineage priors, along with second-order acceleration models, can constrain the inferred paths. For individual motion, researchers combine video, GPS, maps, and sensor data to reconstruct or predict trajectories, with applications in developmental biology, mobility, navigation, and robotics. A central challenge is that many dynamics can fit the observed snapshots, so methods must balance smoothness and physical or biological plausibility with uncertainty, branching, and reliable evaluation when intermediate states are unseen.

Latest papers

37 papers4 papers in the last four weeks, down 20% on the four weeks before. 0.0% of all new papers.

MIOFlow 2.0: A unified framework for inferring cellular stochastic dynamics from single cell and spatial transcriptomics data
Mar 23, 2026Xingzhi Sun, João Felipe Rocha, Brett Phelan +11Trajectory Inference Single-Cell Transcriptomics
Yale University, USA · University of Wisconsin-Madison, USA · Université de Montréal; Mila - Quebec AI Institute, Canada +2
Learning Lineage-guided Geodesics with Finsler Geometry
Mar 17, 2026Aaron Zweig, Mingxuan Zhang, David A. Knowles +1Finsler Geometry Trajectory Inference
New York Genome Center · Irving Institute for Cancer Dynamics, Columbia University · Department of Systems Biology, Columbia University +3
Toward Global Intent Inference for Human Motion by Inverse Reinforcement Learning
Mar 8, 2026Sarmad Mehrdad, Maxime Sabbah, Vincent Bonnet +1Human Motion Prediction Trajectory Inference
Machines in Motion Laboratory, New York University, USA · LAAS-CNRS, Université Paul Sabatier, CNRS, Toulouse, France · Image and Pervasive Access Laboratory (IPAL), CNRS-UMI, Singapore +1
Interface-Aware Trajectory Reconstruction of Limited Demonstrations for Robot Learning
Feb 26, 2026Demiana R. Barsoum, Mahdieh Nejati Javaremi, Larisa Y. C. Loke +1Learning from Suboptimal Demonstrations Trajectory Inference
Northwestern University & Shirley Ryan AbilityLab, Chicago, Illinois, USA
The Swarm Intelligence Freeway-Urban Trajectories (SWIFTraj) Dataset -- Part II: A Graph-Based Approach for Trajectory Connection
Feb 25, 2026Xinkai Ji, Pan Liu, Ying Yang +1Autonomous Driving Datasets Trajectory Inference
School of Transportation, Southeast University, Nanjing, China · School of Management, Shanghai University, Shanghai, China · Department of Architecture and Civil Engineering, Chalmers University of Technology, Gothenburg, Sweden +1
Multi-Marginal Schrödinger Bridge Matching
Oct 18, 2025Byoungwoo Park, Juho LeeSchrödinger Bridges Trajectory Inference
KAIST
Multi-Marginal Flow Matching with Adversarially Learnt Interpolants
Oct 1, 2025Oskar Kviman, Kirill Tamogashev, Nicola Branchini +3Flow Matching Trajectory Inference
KTH · University of Edinburgh

Mar 23, 2026J/K move · Enter open · S save

MIOFlow 2.0: A unified framework for inferring cellular stochastic dynamics from single cell and spatial transcriptomics data

Xingzhi Sun, João Felipe Rocha, Brett Phelan, Dhananjay Bhaskar+10

Yale University, USA · University of Wisconsin-Madison, USA · Université de Montréal; Mila - Quebec AI Institute, Canada+2

Trajectory Inference Single-Cell Transcriptomics Spatial Transcriptomics

Open paper AskPDF arXiv

Understanding cellular trajectories via time-resolved single-cell transcriptomics is vital for studying development, regeneration, and disease. A key challenge is inferring continuous trajectories from discrete snapshots. Biological complexity stems from stochastic cell fate decisions, temporal proliferation changes, and spatial environmental influences. Current methods often use deterministic interpolations treating cells in isolation, failing to capture the probabilistic branching, population shifts, and niche-dependent signaling driving real biological processes. We introduce Manifold Interpolating Optimal-Transport Flow (MIOFlow) 2.0. This framework learns biologically informed cellular trajectories by integrating manifold learning, optimal transport, and neural differential equations. It models three core processes: (1) stochasticity and branching via Neural Stochastic Differential Equations; (2) non-conservative population changes using a learned growth-rate model initialized with unbalanced optimal transport; and (3) environmental influence through a joint latent space unifying gene expression with spatial features like local cell type composition and signaling. By operating in a PHATE-distance matching autoencoder latent space, MIOFlow 2.0 ensures trajectories respect the data's intrinsic geometry. Empirical comparisons show expressive trajectory learning via neural differential equations outperforms existing generative models, including simulation-free flow matching. Validated on synthetic datasets, embryoid body differentiation, and spatially resolved axolotl brain regeneration, MIOFlow 2.0 improves trajectory accuracy and reveals hidden drivers of cellular transitions, like specific signaling niches. MIOFlow 2.0 thus bridges single-cell and spatial transcriptomics to uncover tissue-scale trajectories.