# TALON: Token-Aligned Lightweight Adapters for 6-DoF Spacecraft Pose Estimation

arXiv: https://arxiv.org/abs/2605.31217 · HTML: https://mlsanity.com/paper/2605.31217/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 31 of its 43 references that are in the map.
- First posted on arXiv 29 May 2026.

## Builds on (most cited first, 30 of 31)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [Visual Prompt Tuning](https://mlsanity.com/paper/2203.12119/) (2022)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [Deep High-Resolution Representation Learning for Human Pose Estimation](https://mlsanity.com/paper/1902.09212/) (2019)
- [Vision Transformers Need Registers](https://mlsanity.com/paper/2309.16588/) (2023)
- [AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognition](https://mlsanity.com/paper/2205.13535/) (2022)
- [Deep ViT Features as Dense Visual Descriptors](https://mlsanity.com/paper/2112.05814/) (2021)
- [Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation](https://mlsanity.com/paper/1908.07433/) (2019)
- [ST-Adapter: Parameter-Efficient Image-to-Video Transfer Learning](https://mlsanity.com/paper/2206.13559/) (2022)
- [FoundPose: Unseen Object Pose Estimation with Foundation Features](https://mlsanity.com/paper/2311.18809/) (2023)
- [Segmentation-driven 6D Object Pose Estimation](https://mlsanity.com/paper/1812.02541/) (2018)
- [Numerical Coordinate Regression with Convolutional Neural Networks](https://mlsanity.com/paper/1801.07372/) (2018)
- [SPEED+: Next-Generation Dataset for Spacecraft Pose Estimation across Domain Gap](https://mlsanity.com/paper/2110.03101/) (2021)
- [FreeZe: Training-free zero-shot 6D pose estimation with geometric and vision foundation models](https://mlsanity.com/paper/2312.00947/) (2023)
- [Satellite Pose Estimation with Deep Landmark Regression and Nonlinear Pose Refinement](https://mlsanity.com/paper/1908.11542/) (2019)
- [Deep Learning for Spacecraft Pose Estimation from Photorealistic Rendering](https://mlsanity.com/paper/1907.04298/) (2019)
- [End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames](https://mlsanity.com/paper/2311.17241/) (2023)
- [Wide-Depth-Range 6D Object Pose Estimation in Space](https://mlsanity.com/paper/2104.00337/) (2021)
- [A Survey on Deep Learning-Based Monocular Spacecraft Pose Estimation: Current State, Limitations and Prospects](https://mlsanity.com/paper/2305.07348/) (2023)
- [Pose Estimation for Non-Cooperative Rendezvous Using Neural Networks](https://mlsanity.com/paper/1906.09868/) (2019)
- [SPADES: A Realistic Spacecraft Pose Estimation Dataset using Event Sensing](https://mlsanity.com/paper/2311.05310/) (2023)
- [ChiNet: Deep Recurrent Convolutional Learning for Multimodal Spacecraft Pose Estimation](https://mlsanity.com/paper/2108.10282/) (2021)
- [Efficient Transfer Learning for Video-language Foundation Models](https://mlsanity.com/paper/2411.11223/) (2024)
- [FastPose-ViT: A Vision Transformer for Real-Time Spacecraft Pose Estimation](https://mlsanity.com/paper/2512.09792/) (2025)
- [Motion Aware ViT-based Framework for Monocular 6-DoF Spacecraft Pose Estimation](https://mlsanity.com/paper/2509.06000/) (2025)

## Abstract (arXiv)

Monocular 6-DoF spacecraft pose estimation methods predominantly process individual frames, discarding the temporal information present in an image sequence acquired during spacecraft manoeuvres. Few temporal approaches require full backbone fine-tuning or auxiliary optical flow networks, risking catastrophic forgetting or increasing computational cost, respectively. We propose TALON (Token-Aligned Lightweight adapters for Orbital Navigation): spatiotemporal 3D adapters injected before the self-attention layers of a frozen ViT vision transformer, combined with a patch-token alignment loss that geometrically grounds the adapted features to keypoint structure through a prototype-conditioned KL-divergence objective. Pre-attention placement allows the frozen attention to reason over temporally enriched tokens, achieving stronger performance with a single adapter per block than post-attention alternatives. The alignment loss shapes the intermediate representations so that each keypoint induces a spatially precise activation in the token field, while the framework adds less than 5% parameters to the frozen backbone. On SPADES dataset, TALON reduces the pose error by 50% over the prior state-of-the-art, and on SwissCube dataset it surpasses the prior best by 21.8% in ADD-0.1d accuracy. Zero-shot cross-domain evaluation from sim-to-real on SPARK real data reduces pose error by 4.7x, and ablations characterise the role of adapter depth across in-domain and cross-domain settings.