Act3D: 3D Feature Field Transformers for Multi-Task Robotic Manipulation
read the original abstract
3D perceptual representations are well suited for robot manipulation as they easily encode occlusions and simplify spatial reasoning. Many manipulation tasks require high spatial precision in end-effector pose prediction, which typically demands high-resolution 3D feature grids that are computationally expensive to process. As a result, most manipulation policies operate directly in 2D, foregoing 3D inductive biases. In this paper, we introduce Act3D, a manipulation policy transformer that represents the robot's workspace using a 3D feature field with adaptive resolutions dependent on the task at hand. The model lifts 2D pre-trained features to 3D using sensed depth, and attends to them to compute features for sampled 3D points. It samples 3D point grids in a coarse to fine manner, featurizes them using relative-position attention, and selects where to focus the next round of point sampling. In this way, it efficiently computes 3D action maps of high spatial resolution. Act3D sets a new state-of-the-art in RL-Bench, an established manipulation benchmark, where it achieves 10% absolute improvement over the previous SOTA 2D multi-view policy on 74 RLBench tasks and 22% absolute improvement with 3x less compute over the previous SOTA 3D policy. We quantify the importance of relative spatial attention, large-scale vision-language pre-trained 2D backbones, and weight tying across coarse-to-fine attentions in ablative experiments. Code and videos are available on our project website: https://act3d.github.io/.
This paper has not been read by Pith yet.
Forward citations
Cited by 14 Pith papers
-
Learning to See While Learning to Act: Diffusion Models for Active Perception in Robot Imitation
See2Act couples action denoising with viewpoint refinement in a diffusion-based imitation learning policy trained on keyframe-anchored camera poses, recovering informative views under occlusion and improving RLBench p...
-
From Grasps to Dexterity: Large-Scale Grasp Pretraining for Dexterous Manipulation
Grasp pretraining on 355k trajectories improves full-task success on six articulated tool-use tasks by 33.3 pp over DP3 in real-world experiments.
-
SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation
SSI-Policy uses an RGB-only Structured Scene Interface to improve LIBERO benchmark performance by nearly 15% with only 10 demonstrations per task compared to prior methods.
-
SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation
SSI-Policy learns a robot-agnostic RGB-only scene interface from video to improve vision-language manipulation policies by 15% on LIBERO with only 10 demos per task.
-
Fourier Features Let Agents Learn High Precision Policies with Imitation Learning
Mapping point clouds to Fourier features improves high-precision imitation learning policies on RoboCasa, ManiSkill3, and real-robot tasks compared with Cartesian inputs.
-
Multi-View Video Diffusion Policy: A 3D Spatio-Temporal-Aware Video Action Model
MV-VDP jointly predicts multi-view RGB and heatmap videos via diffusion to achieve data-efficient, robust robotic manipulation policies.
-
GAF: Gaussian Action Field as a 4D Representation for Dynamic World Modeling in Robotic Manipulation
GAF creates 4D dynamic scene models by adding motion to 3D Gaussians, enabling better reconstruction and 7.3% higher success in robotic tasks.
-
3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations
DP3 uses compact 3D representations from sparse point clouds inside diffusion policies to learn generalizable visuomotor skills from few demonstrations, reporting 24% gains in simulation and 85% success on real robots.
-
LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories
LabVLA uses RoboGenesis simulation data and a two-stage FAST pretraining plus flow matching recipe on a Qwen3-VL backbone to achieve the highest success rates on LabUtopia under in- and out-of-distribution conditions.
-
OASIS: Observation-Action Space Alignment via SE(3) Trajectory Prediction for Robotic Manipulation
OASIS improves robotic manipulation success and generalization by predicting camera-frame SE(3) end-effector trajectories to condition the action decoder on pose-supervised states.
-
Learning Structural Latent Points for Efficient Visual Representations in Robotic Manipulation
A hybrid structural latent points representation is learned by inserting a point-wise latent VAE into a point-cloud autoencoder and regularizing toward a Gaussian prior, paired with a lightweight 3DGS rendering pipeli...
-
X-Imitator: Spatial-Aware Imitation Learning via Bidirectional Action-Pose Interaction
X-Imitator is a bidirectional action-pose interaction framework for spatial-aware imitation learning that outperforms vanilla policies and explicit pose guidance on 24 simulated and 3 real-world robotic tasks.
-
Uni-Encoder Meets Multi-Encoders: Representation Before Fusion for Brain Tumor Segmentation with Missing Modalities
UniME combines a pretrained unified ViT encoder with modality-specific CNN encoders to improve brain tumor segmentation performance when some MRI modalities are missing.
-
R3D: Revisiting 3D Policy Learning
A transformer 3D encoder plus diffusion decoder architecture, with 3D-specific augmentations, outperforms prior 3D policy methods on manipulation benchmarks by improving training stability.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.