# WildBox: A Dataset and Benchmark for Aerial Monocular 3D Detection of African Savanna Wildlife

arXiv: https://arxiv.org/abs/2606.21309 · HTML: https://mlsanity.com/paper/2606.21309/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 51 references that are in the map.
- First posted on arXiv 19 Jun 2026.

## Builds on (most cited first, 29 of 29)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [Argoverse: 3D Tracking and Forecasting with Rich Maps](https://mlsanity.com/paper/1911.02620/) (2019)
- [Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding](https://mlsanity.com/paper/2011.02523/) (2020)
- [ARKitScenes: A Diverse Real-World Dataset For 3D Indoor Scene Understanding Using Mobile RGB-D Data](https://mlsanity.com/paper/2111.08897/) (2021)
- [UniDepth: Universal Monocular Metric Depth Estimation](https://mlsanity.com/paper/2403.18913/) (2024)
- [Omni3D: A Large Benchmark and Model for 3D Object Detection in the Wild](https://mlsanity.com/paper/2207.10660/) (2022)
- [Objectron: A Large Scale Dataset of Object-Centric Videos in the Wild with Pose Annotations](https://mlsanity.com/paper/2012.09988/) (2020)
- [Seeing biodiversity: perspectives in machine learning for wildlife conservation](https://mlsanity.com/paper/2110.12951/) (2021)
- [Three-D Safari: Learning to Estimate Zebra Pose, Shape, and Texture from Images "In the Wild"](https://mlsanity.com/paper/1908.07201/) (2019)
- [Animal3D: A Comprehensive Dataset of 3D Animal Pose and Shape](https://mlsanity.com/paper/2308.11737/) (2023)
- [Detect Anything 3D in the Wild](https://mlsanity.com/paper/2504.07958/) (2025)
- [AcinoSet: A 3D Pose Estimation Dataset and Baseline Models for Cheetahs in the Wild](https://mlsanity.com/paper/2103.13282/) (2021)
- [Open Vocabulary Monocular 3D Object Detection](https://mlsanity.com/paper/2411.16833/) (2024)
- [3D-POP -- An automated annotation approach to facilitate markerless 2D-3D tracking of freely moving birds with marker-based motion capture](https://mlsanity.com/paper/2303.13174/) (2023)
- [BaboonLand Dataset: Tracking Primates in the Wild and Automating Behaviour Recognition from Drone Videos](https://mlsanity.com/paper/2405.17698/) (2024)
- [BuckTales : A multi-UAV dataset for multi-object tracking and re-identification of wild antelopes](https://mlsanity.com/paper/2411.06896/) (2024)
- [GRADE: Generating Realistic And Dynamic Environments for Robotics Research with Isaac Sim](https://mlsanity.com/paper/2303.04466/) (2023)
- [Synthetic Data-based Detection of Zebras in Drone Imagery](https://mlsanity.com/paper/2305.00432/) (2023)
- [LabelAny3D: Label Any Object 3D in the Wild](https://mlsanity.com/paper/2601.01676/) (2026)
- [POLO -- Point-based, multi-class animal detection](https://mlsanity.com/paper/2410.11741/) (2024)
- [WildLive: Near Real-time Visual Wildlife Tracking onboard UAVs](https://mlsanity.com/paper/2504.10165/) (2025)
- [MOO: A Multi-view Oriented Observations Dataset for Viewpoint Analysis in Cattle Re-Identification](https://mlsanity.com/paper/2603.04314/) (2026)
- [Adapting the re-ID challenge for static sensors](https://mlsanity.com/paper/2412.00290/) (2024)
- [WildLIFT: Lifting monocular drone video to 3D for species-agnostic wildlife monitoring](https://mlsanity.com/paper/2604.24718/) (2026)

## Abstract (arXiv)

We introduce WildBox, a dataset and benchmark for monocular 3D detection of wildlife from drone video, comprising 237,505 3D bounding box annotations across seven African savanna species grouped into six benchmark classes. Annotations follow a KITTI/Omni3D-compatible format in a per-segment scale-normalised camera frame, with instance identities maintained across each segment. We evaluate two open-vocabulary monocular 3D architectures, OVMono3D-LIFT and DetAny3D, under zero-shot, ground-truth 2D box prompt, and supervised fine-tuning protocols. Open-vocabulary 2D foundation models provide usable zero-shot wildlife localisation (50.55 AP@50), but zero-shot 3D detection collapses to 0.00 AP across both architectures and every 2D-input condition tested, including ground-truth 2D box prompts, thus isolating the failure to the 3D stage. Fine-tuning on WildBox recovers performance to 8.68 +/- 0.47 AP-BEV@0.50 and 13.17 +/- 0.69 AP3D macro. Depth contributes 84% of normalised Hausdorff distance after fine-tuning and over 99% in zero-shot, identifying monocular aerial depth as the dominant open problem in this regime. A coarse-to-fine curriculum, i.e. pretraining on a merged zebra class before fine-tuning on the Grevy's/plains split, improves macro 3D performance with less total compute, with the largest gains on the two zebra subclasses. WildBox is released with video-level splits, evaluation code, and baseline checkpoints to enable progress in 3D wildlife perception from drone video.