MapleGrasp: Mask-guided Feature Pooling for Language-driven Efficient Robotic Grasping (WACV 2026)

Authors: Vineet Bhat, Naman Patel, Prashanth Krishnamurthy, Ramesh Karri and Farshad Khorrami

This is an official PyTorch implementation of the baseline end-to-end model MapleGrasp of our work. The implementation of our MapleGrasp model is based on the CROG model, thanks for their amazing work!

Robotic manipulation of unseen objects via natural language commands remains challenging. Language driven robotic grasping (LDRG) predicts stable grasp poses from natural language queries and RGB-D images. We propose MapleGrasp, a novel framework that leverages mask-guided feature pooling for efficient vision-language driven grasping. Our two-stage training first predicts segmentation masks from CLIP-based vision-language features. The second stage pools features within these masks to generate pixel-level grasp predictions, improving efficiency, and reducing computation. Incorporating mask pooling results in a 7% improvement over prior approaches on the OCID-VLG benchmark. Furthermore, we introduce RefGraspNet, an open-source dataset eight times larger than existing alternatives, significantly enhancing model generalization for open-vocabulary grasping. MapleGrasp scores a strong grasping accuracy of 89% when compared with competing methods in the RefGraspNet benchmark. Our method achieves comparable performance to larger Vision-Language-Action models on the LIBERO benchmark, and shows significantly better generalization to unseen tasks. Real-world experiments on a Franka arm demonstrate 73% success rate with unseen objects, surpassing competitive baselines by 11%.

Preparation

1. Create conda environment

      conda env create -f environment.yml

2. Datasets
   • Please download the OCID-VLG dataset by following the instructions and directory formatting.

Training

python train_maplegrasp.py --config config/OCID-VLG/maplegrasp_multiple_r50.yaml

Please appropriately modify the stage 1 and stage 2 parameters in the yaml file before running.

Evaluation

python test.py --config config/OCID-VLG/maplegrasp_multiple_r50.yaml

Please remember to modify the path to the dataset in config files.

License

This project is under the MIT license. See LICENSE for details.

Citation

If you find our work useful in your research, please consider citing:

@article{bhat2025maplegrasp,
  title={MapleGrasp: Mask-guided Feature Pooling for Language-driven Efficient Robotic Grasping},
  author={Bhat, Vineet and Patel, Naman and Krishnamurthy, Prashanth and Karri, Ramesh and Khorrami, Farshad},
  journal={arXiv preprint arXiv:2506.06535},
  year={2025}
}