TY - GEN
T1 - FutureHuman3D
T2 - 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024
AU - Diller, Christian
AU - Funkhouser, Thomas
AU - Dai, Angela
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - We present a generative approach to forecast long-term future human behavior in 3D, requiring only weak supervision from readily available 2D human action data. This is a fundamental task enabling many downstream applications. The required ground-truth data is hard to capture in 3D (mocap suits, expensive setups) but easy to acquire in 2D (simple RGB cameras). Thus, we design our method to only require 2D RGB data at inference time while being able to generate 3D human motion sequences. We use a differentiable 2D projection scheme in an autoregressive manner for weak supervision, and an adversarial loss for 3D regularization. Our method predicts long and complex human behavior sequences (e.g., cooking, assembly) consisting of multiple sub-actions. We tackle this in a semantically hierarchical manner, jointly predicting high-level coarse action labels together with their low-level fine-grained realizations as characteristic 3D human poses. We observe that these two action representations are coupled in nature, and joint prediction benefits both action and pose forecasting. Our experiments demonstrate the complementary nature of joint action and 3D pose prediction: our joint approach outperforms each task treated individually, enables robust longer-term sequence prediction, and improves over alternative approaches to forecast actions and characteristic 3D poses.
AB - We present a generative approach to forecast long-term future human behavior in 3D, requiring only weak supervision from readily available 2D human action data. This is a fundamental task enabling many downstream applications. The required ground-truth data is hard to capture in 3D (mocap suits, expensive setups) but easy to acquire in 2D (simple RGB cameras). Thus, we design our method to only require 2D RGB data at inference time while being able to generate 3D human motion sequences. We use a differentiable 2D projection scheme in an autoregressive manner for weak supervision, and an adversarial loss for 3D regularization. Our method predicts long and complex human behavior sequences (e.g., cooking, assembly) consisting of multiple sub-actions. We tackle this in a semantically hierarchical manner, jointly predicting high-level coarse action labels together with their low-level fine-grained realizations as characteristic 3D human poses. We observe that these two action representations are coupled in nature, and joint prediction benefits both action and pose forecasting. Our experiments demonstrate the complementary nature of joint action and 3D pose prediction: our joint approach outperforms each task treated individually, enables robust longer-term sequence prediction, and improves over alternative approaches to forecast actions and characteristic 3D poses.
UR - https://www.scopus.com/pages/publications/85205822715
U2 - 10.1109/CVPR52733.2024.01881
DO - 10.1109/CVPR52733.2024.01881
M3 - Conference contribution
AN - SCOPUS:85205822715
SN - 9798350353006
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 19902
EP - 19914
BT - Proceedings - 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024
PB - IEEE Computer Society
Y2 - 16 June 2024 through 22 June 2024
ER -