# Don't waste SAM

arXiv: https://arxiv.org/abs/2606.10696 · HTML: https://mlsanity.com/paper/2606.10696/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 9 Jun 2026.

## Builds on (most cited first, 9 of 9)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Reviving Iterative Training with Mask Guidance for Interactive Segmentation](https://mlsanity.com/paper/2102.06583/) (2021)
- [Anything-3D: Towards Single-view Anything Reconstruction in the Wild](https://mlsanity.com/paper/2304.10261/) (2023)
- [TrashCan: A Semantically-Segmented Dataset towards Visual Detection of Marine Debris](https://mlsanity.com/paper/2007.08097/) (2020)
- [TACO: Trash Annotations in Context for Litter Detection](https://mlsanity.com/paper/2003.06975/) (2020)
- [Any-to-Any Style Transfer: Making Picasso and Da Vinci Collaborate](https://mlsanity.com/paper/2304.09728/) (2023)
- [Learning to "Segment Anything" in Thermal Infrared Images through Knowledge Distillation with a Large Scale Dataset SATIR](https://mlsanity.com/paper/2304.07969/) (2023)

## Abstract (arXiv)

Meta AI has recently released the Segment Anything Model (SAM), which demonstrates exceptional zero-shot image segmentation performance across various tasks with remarkable accuracy. Despite its inability to provide accurate segmentation across multiple research fields, SAM still serves as a valuable starting point for supporting the segmentation pipeline process, particularly for tasks that require extensive and senior skills annotations. This study aims to evaluate the generalization of SAM and fine-tuning SAM models using three waste segmentation datasets. Although they are captured from real scenes as SAM was pretrained on, these datasets present several challenges, including occlusions, deformable objects, transparency, and objects easily confused with backgrounds. In our findings, the fine-tuned SAM-ViT-H model outperforms the state-ofthe-art Zerowaste, and TACO datasets with a significant increase of +30 in IoU, and it closely approaches performance levels of TrashCan 1.0, with only a -1.44 difference. After evaluating these popular waste datasets, it became evident that fine-tuning SAM as a foundational model is a crucial step for providing better generalization for downstream waste segmentation tasks. Therefore, SAM should not be disregarded or wasted.