Language Grounding with 3D Objects
By Jesse Thomason, Mohit Shridhar, Yonatan Bisk, Chris Paxton, Luke S. Zettlemoyer. 2021
56 citations recorded by the source corpus.
Seemingly simple natural language requests to a robot are generally underspecified, for example"Can you bring me the wireless mouse?"Flat images of candidate mice may not provide the discriminative information needed for"wireless."The world, and objects in it, are not flat images but complex 3D shapes. If a human requests an object based on any of its basic properties, such as color, shape, or texture, robots should perform the necessary exploration to accomplish the task. In particular, while substantial effort and progress has been made on understanding explicitly visual attributes like color and category, comparatively little progress has been made on understanding language about shapes and contours. In this work, we introduce a novel reasoning task that targets both visual and non-visual language about 3D objects. Our new benchmark, ShapeNet Annotated with Referring Expressions (SNARE) requires a model to choose which of two objects is being referenced by a natural language description. We introduce several CLIP-based models for distinguishing objects and demonstrate that while recent advances in jointly modeling vision and language are useful for robotic language understanding, it is still the case that these image-based models are weaker at understanding the 3D nature of objects -- properties which play a key role in manipulation. We find that adding view estimation to language grounding models improves accuracy on both SNARE and when identifying objects referred to in language on a robot platform, but note that a large gap remains between these models and human performance.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sergey Levine
- Chelsea Finn
- Pieter Abbeel
- Trevor Darrell
- Abhinav Kumar Gupta
- Mohit Bansal
- Li Fei-Fei
- Quoc V. Le
In Computer Vision and Pattern Recognition
- Sergey Levine
- Luc Van Gool
- Jitendra Malik
- Dacheng Tao
- Trevor Darrell
- Li Fei-Fei
- Andrew Zisserman
- Xiaogang Wang
In Natural Language Processing
- Antonio Torralba
- Sergey Levine
- Dieter Fox
- Jacob Andreas
- Chelsea Finn
- Dorsa Sadigh
- Pratyusha Sharma
- Brian Ichter
Citations in the corpus
- Do As I Can, Not As I Say: Grounding Language in Robotic Affordances (2022, background): “Recent works have studied how to ground modern language models, by training them to accept additional environment inputs [23, 24, 25, 26, 27, 28, 29] or to directly output actions [30, 31, 32].”
- LERF: Language Embedded Radiance Fields (2023, background): “In addition, incorporating language with shape information can improve object recognition via text [11, 36].”
- Correcting Robot Plans with Natural Language Feedback (2022, methodology): “The CLIP embeddings used in our model provide a strong basis for language generalization, as seen in previous work [40, 35].”
- SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities (2024, background): “Others have focused on fine-grained scene understanding, such as semantic segmentation [5, 35], object detection [11], or object identification [15, 57].”