Congratulations to the awardee and all finalists of the Rising Star Award for Spatial Intelligence!
| Name | Institute | Stage | Homepage |
|---|---|---|---|
| 🏆 Yining Hong | Stanford | Postdoc | evelinehong.github.io |
In alphabetical order:
| Name | Institute | Stage | Homepage |
|---|---|---|---|
| Zhiyang Dou | MIT | PhD Student | people.csail.mit.edu/frankzydou |
| Jiafei Duan | University of Washington | PhD Student | jiafei1224.github.io |
| Hezhen Hu | UT Austin | Postdoc | alexhu.top |
| Tiange Xiang | Stanford | PhD Student | ai.stanford.edu/~xtiange |
| Junyi Zhang | UC Berkeley | PhD Student | junyi42.com |
Both awards are generously provided by 2077AI: up to USD 30,000 in research gift funding to the awardee's institution, and USD 2,000 in API quota to each finalist on the PPAPI platform.
We sincerely appreciate every applicant's participation. As a thank-you, each applicant is eligible for 200 credits on the PPAPI platform to support their own research projects. To claim your credits, please email contact@2077ai.com.
We sincerely thank our Senior Advisory Committee for their careful review and guidance throughout the selection process.
| Proposal Submission Deadline | May 24, 2026, 11:59 PM PST |
|---|---|
| Announcement | Due to a high volume of applications, the announcement will be made on June 2. |
| Workshop Date | June 3, 2026, CVPR 2026, Denver |
| Selection | 1 final awardee and 5 finalists |
| Eligibility | Current PhD students and postdoctoral researchers |
| Research Gift Fund | Up to USD 30,000 to the institution of the awardee as gift funding. No indirect cost is intended. Final terms are subject to sponsor approval and institutional policies. |
The Rising Star Award for Spatial Intelligence is organized as part of the E2E3D Workshop at CVPR 2026. The award recognizes an early-career researcher with strong research achievements and a clear future vision for spatial intelligence.
The award focuses on work that connects 3D perception, 3D generation, 3D representations, world models, spatial reasoning, embodied AI, XR, and real-time systems. The goal is to highlight a research agenda that helps future models understand, reconstruct, generate, and act in 3D and dynamic environments.
Applicants must be in one of the following roles at the time of submission:
Additional guidelines:
Each applicant should submit three PDF documents. The Research Achievement Summary and Research Vision Proposal should each be at most one page; the Advisor Reference Letter has no page limit.
Applicants may use any format, but a strong one-page proposal should state the core research question, explain why the problem matters now, outline the technical path, define how progress will be measured, and describe what the community will gain.
Relevant areas include, but are not limited to:
| Area | Example Topics |
|---|---|
| 3D/4D reconstruction | Single-view, multi-view, and video-to-3D/4D reconstruction; RGB-D and LiDAR reconstruction; SLAM, mapping, localization, and geometry from unposed images. |
| 3D/4D generation and editing | Object, scene, human, and asset generation; text-to-3D/4D, image-to-3D/4D, video-to-3D/4D; controllable editing; physically grounded generation. |
| 3D/4D representations | Neural radiance fields, 3D Gaussian splatting, implicit fields, meshes, point clouds, occupancy fields, signed distance fields, and hybrid representations. |
| Video generation and world models | Dynamic scenes, temporal consistency, persistent spatial memory, physical prediction, scene simulation, and long-horizon video understanding. |
| Spatial reasoning | 3D vision-language models, spatial relation understanding, scene graphs, layout reasoning, geometric and physical reasoning, and action-conditioned perception. |
| Embodied AI | Vision-language-action models, world-action models, navigation, manipulation, affordance learning, policy learning in 3D environments, and sim-to-real transfer. |
| Spatial computing, XR, and mixed reality | AR, VR, mixed reality, spatial maps, real-time scene understanding, user-facing 3D perception, and interaction in spatial computing systems. |
| Data, evaluation, and systems | Large-scale 3D and video data engines, synthetic data, multi-sensor logs, benchmarks, robustness, latency, memory, energy, and edge deployment. |
Applications will be reviewed by the award committee. Reviewers will consider the career stage of the applicant. Main criteria include:
This call may be updated on the workshop website. Applicants should follow the final instructions posted on the E2E3D website.
Trouble loading the form below? Open it directly: https://forms.gle/2Vm7mtZWKejG85oG6
You will need to be signed in to a Google account to upload PDFs.