Abstract
In this paper, we address the Cross-View Object Re-Identification (ReID) task, aiming to match identities across drastically different viewpoints. We focus specifically on Single-View Aerial-Ground ReID (SV AG-ReID), where models train on one viewpoint and must generalize to another at inference. To bridge the viewpoint gap of Cross-View ReID, related work employs generative models to synthesize novel views from source images as additional training data. We leverage recent advances in large-scale single-view 3D mesh reconstruction to lift source images into textured meshes and to render them from new viewpoints. However, this pipeline produces unreliable data in two distinct ways, and addressing both is vital for synthetic data to serve as an effective training signal. First, the reconstruction may fail and produce geometrically incorrect meshes, resulting in severe corruption in rendered novel views. Second, even when the geometry is correct, the rendered novel views contain hallucinated textures in regions not visible in the source image, which may degrade training if used naively. We address both shortcomings by exploiting a signal that prior generative methods do not naturally expose. Indeed, the reconstructed mesh carries a per-pixel partition between regions visible from the source view and regions that the reconstruction model had to hallucinate, available as a deterministic byproduct of rendering. We introduce this per-pixel visibility map and exploit it in a two-stage design. Cross-View Consistency Filtering (CVCF) discards geometrically incorrect meshes by ensuring multi-view consistency between source and novel views over visible patches. Visibility-Aware Attention Supervision (VAAS) leverages the visibility map to guide the model’s attention toward reliable regions of the rendered view during training. Combined, these two components enable our method, 3D-VITAL, to better leverage the synthetic signal from single-view 3D mesh reconstruction, benefiting from wider viewpoint augmentation. 3D-VITAL sets a new state-of-the-art on three SV AG-ReID benchmarks (AG-ReID, AG-ReID.v2, MOO).
The code is available at https://github.com/TurtleSmoke/3D-VITAL, and the checkpoints on Hugging Face at TurtleSmoke/3D-VITAL.
Method
3D-VITAL is built upon 3D-LENS and extends it with three components that improve the use of synthetic data from single-view 3D mesh reconstruction. First, we compute a per-pixel visibility map that separates visible regions from hallucinated regions. Second, Cross-View Consistency Filtering (CVCF) uses this visibility map to reject geometrically incorrect samples. Third, Visibility-Aware Attention Supervision (VAAS) operates on the surviving samples and uses the visibility map both as a per-patch attention target and as a per-sample weight.
Results
By filtering broken meshes and steering the attention toward the observed regions, 3D-VITAL can exploit a wider viewpoint augmentation than 3D-LENS and improves over it on AG-ReID and AG-ReID.v2.
Citation
@inproceedings{grolleau2026vital, author = {Grolleau, William and Sabourin, Astrid and Lapouge, Guillaume and Achard, Catherine}, title = {{3D-VITAL}: {V}isibility-aware Identity Training from Artificial Liftings of {3D} Reconstruction model for Cross-View Object Re-Identification}, booktitle = {Advances in Neural Information Processing Systems (NeurIPS)}, year = {2026},}