acceptodds
Withdrawn from ICLR 2027

Embodied Drone Tracking in the Wild: Benchmark and Baseline

Withdrawn from ICLR 2027, so left out of the paper lists and the map.

Abstract

Embodied drone tracking requires an agent to follow a moving target while actively controlling its viewpoint and avoiding obstacles. Drones offer broad views and flexible altitude adjustment, yet target motion and interrupted visibility make this coupling challenging. Existing benchmarks provide limited coverage of substantial target elevation changes and complex occlusions under a common promptable tracking protocol. We introduce the Embodied Drone Tracking Benchmark (EDT-BENCH), built on UnrealZoo to address this gap. It supports language, point, bounding box, mask, and template initialization across diverse environments. The training corpus contains 5,670 expert trajectories and their perturbed counterparts from 81 scenes, providing aligned supervision for tracking and recovery. We further propose EDT, a simple baseline built on frozen SAM2 representations. Its action head estimates relative target pose from current observations and temporal memory, then uses uncertainty-aware fusion to predict drone waypoints. The memory branch retains target information during occlusion, while spatial prompts reduce ambiguity when similar distractors are present. A unified closed-loop protocol evaluates tracking robustness and generalization across scenes and target categories. Together, EDT-BENCH and EDT provide a reproducible closed-loop framework for evaluating prompt-conditioned tracking, spatial tracking accuracy, recovery from occlusion and viewpoint perturbations, and generalization across scenes and target categories.

Primary area:
datasets and benchmarks
Keywords:
Embodied drone tracking, multi-prompt tracking, SAM2, imitation learning

Reviews

Among:
Shape:
1234this paper, Bottom 1%
Figure 1. Mean review rating of the reviewed papers at ICLR 2027. This paper averages 1.00 over 3 reviews, above 0% of the others.
Reviewer_3U7rOpenReview ↗
Rating
1
Confidence
4
Informativeness
1
Presentation
1
Soundness
1
Read review
Paper summary

This paper is an imcomplete paper, with several "-" in experiment table.

Critical strengths

None

Critical weaknesses

This paper is more likely a AI generated paper, with AI style figure and incomplete experiment result.

Questions for authors

N/A

Reviewer_GWrCOpenReview ↗
Rating
1
Confidence
4
Informativeness
1
Presentation
1
Soundness
1
Read review
Paper summary

I strongly recommend rejection. This paper should never have been submitted in its current state. It contains no experimental results whatsoever and is clearly an unfinished piece of work. Submitting incomplete work like this wastes reviewers' time and undermines the peer review process.

Critical strengths

NA

Critical weaknesses

I strongly recommend rejection. This paper should never have been submitted in its current state. It contains no experimental results whatsoever and is clearly an unfinished piece of work. Submitting incomplete work like this wastes reviewers' time and undermines the peer review process.

Questions for authors

NA

Reviewer_DryrOpenReview ↗
Rating
1
Confidence
4
Informativeness
1
Presentation
2
Soundness
1
Read review
Paper summary

A simulated drone-tracking benchmark and SAM2 baseline, with no closed-loop results.

Critical strengths

The manuscript is formally complete as a paper.

Critical weaknesses

No results provided (Table 4 / 5 / 6). Although Section 5.1, “Status of numerical evaluation,” explains why Table 4 is filled with dashes “Dashes in the tables denote pending measurements, not zero scores", the paper still falls well short of the ICLR bar.

Questions for authors

N/A

From OpenReview, CC BY 4.0. Authors as signed there; shown unedited.

Related papers

Discussion (0)

Sign in to comment.