Max Hamilton
@max-ham.bsky.social
👾CV/ML PhD Student at UMass Amherst
The text comes from Wikipedia articles of each species, and the observation data is from iNaturalist, same as SINR (arxiv.org/abs/2306.02564). Because of the observation data, predictions are biased toward regions enclosed by actual species boundaries, making human-made borders tricky
December 9, 2024 at 10:31 PM
The text comes from Wikipedia articles of each species, and the observation data is from iNaturalist, same as SINR (arxiv.org/abs/2306.02564). Because of the observation data, predictions are biased toward regions enclosed by actual species boundaries, making human-made borders tricky
🔗Check out the our paper and code:
Paper: arxiv.org/abs/2410.10931
Code: github.com/cvl-umass/le...
If you'll be attending #NeurIPS2024 this week, stop by our poster to learn more!
December 13th, 11am-2pm, East Exhibit Hall A-C #3903
Paper: arxiv.org/abs/2410.10931
Code: github.com/cvl-umass/le...
If you'll be attending #NeurIPS2024 this week, stop by our poster to learn more!
December 13th, 11am-2pm, East Exhibit Hall A-C #3903
Combining Observational Data and Language for Species Range Estimation
Species range maps (SRMs) are essential tools for research and policy-making in ecology, conservation, and environmental management. However, traditional SRMs rely on the availability of environmental...
arxiv.org
December 9, 2024 at 3:13 PM
🔗Check out the our paper and code:
Paper: arxiv.org/abs/2410.10931
Code: github.com/cvl-umass/le...
If you'll be attending #NeurIPS2024 this week, stop by our poster to learn more!
December 13th, 11am-2pm, East Exhibit Hall A-C #3903
Paper: arxiv.org/abs/2410.10931
Code: github.com/cvl-umass/le...
If you'll be attending #NeurIPS2024 this week, stop by our poster to learn more!
December 13th, 11am-2pm, East Exhibit Hall A-C #3903
🌎LE-SINR is able to geographically ground text prompts to locations on the earth. This includes continents and countries, geographic features, and even concepts that do not appear in our species training data but are represented in the language model 🔍Example: "hello kitty"
December 9, 2024 at 3:13 PM
🌎LE-SINR is able to geographically ground text prompts to locations on the earth. This includes continents and countries, geographic features, and even concepts that do not appear in our species training data but are represented in the language model 🔍Example: "hello kitty"
🚀Le-SINR combines millions of citizen science species observations with textual descriptions from Wikipedia, enabling zero-shot range estimation from text and facilitating the learning of rich spatial covariates at a global scale.
Low-dimensional projection of the learned features 👇
Low-dimensional projection of the learned features 👇
December 9, 2024 at 3:12 PM
🚀Le-SINR combines millions of citizen science species observations with textual descriptions from Wikipedia, enabling zero-shot range estimation from text and facilitating the learning of rich spatial covariates at a global scale.
Low-dimensional projection of the learned features 👇
Low-dimensional projection of the learned features 👇