Dyna-DepthFormer: Multi-frame Transformer for Self-Supervised Depth
Estimation in Dynamic Scenes
- URL: http://arxiv.org/abs/2301.05871v2
- Date: Mon, 20 Mar 2023 09:12:21 GMT
- Title: Dyna-DepthFormer: Multi-frame Transformer for Self-Supervised Depth
Estimation in Dynamic Scenes
- Authors: Songchun Zhang and Chunhui Zhao
- Abstract summary: We propose a novel Dyna-Depthformer framework, which predicts scene depth and 3D motion field jointly.
Our contributions are two-fold. First, we leverage multi-view correlation through a series of self- and cross-attention layers in order to obtain enhanced depth feature representation.
Second, we propose a warping-based Motion Network to estimate the motion field of dynamic objects without using semantic prior.
- Score: 19.810725397641406
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Self-supervised methods have showed promising results on depth estimation
task. However, previous methods estimate the target depth map and camera
ego-motion simultaneously, underusing multi-frame correlation information and
ignoring the motion of dynamic objects. In this paper, we propose a novel
Dyna-Depthformer framework, which predicts scene depth and 3D motion field
jointly and aggregates multi-frame information with transformer. Our
contributions are two-fold. First, we leverage multi-view correlation through a
series of self- and cross-attention layers in order to obtain enhanced depth
feature representation. Specifically, we use the perspective transformation to
acquire the initial reference point, and use deformable attention to reduce the
computational cost. Second, we propose a warping-based Motion Network to
estimate the motion field of dynamic objects without using semantic prior. To
improve the motion field predictions, we propose an iterative optimization
strategy, together with a sparsity-regularized loss. The entire pipeline
achieves end-to-end self-supervised training by constructing a minimum
reprojection loss. Extensive experiments on the KITTI and Cityscapes benchmarks
demonstrate the effectiveness of our method and show that our method
outperforms state-of-the-art algorithms.
Related papers
- MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion [118.74385965694694]
We present Motion DUSt3R (MonST3R), a novel geometry-first approach that directly estimates per-timestep geometry from dynamic scenes.
By simply estimating a pointmap for each timestep, we can effectively adapt DUST3R's representation, previously only used for static scenes, to dynamic scenes.
We show that by posing the problem as a fine-tuning task, identifying several suitable datasets, and strategically training the model on this limited data, we can surprisingly enable the model to handle dynamics.
arXiv Detail & Related papers (2024-10-04T18:00:07Z) - Let Occ Flow: Self-Supervised 3D Occupancy Flow Prediction [14.866463843514156]
Let Occ Flow is the first self-supervised work for joint 3D occupancy and occupancy flow prediction using only camera inputs.
Our approach incorporates a novel attention-based temporal fusion module to capture dynamic object dependencies.
Our method extends differentiable rendering to 3D volumetric flow fields.
arXiv Detail & Related papers (2024-07-10T12:20:11Z) - Motion-aware 3D Gaussian Splatting for Efficient Dynamic Scene Reconstruction [89.53963284958037]
We propose a novel motion-aware enhancement framework for dynamic scene reconstruction.
Specifically, we first establish a correspondence between 3D Gaussian movements and pixel-level flow.
For the prevalent deformation-based paradigm that presents a harder optimization problem, a transient-aware deformation auxiliary module is proposed.
arXiv Detail & Related papers (2024-03-18T03:46:26Z) - DO3D: Self-supervised Learning of Decomposed Object-aware 3D Motion and
Depth from Monocular Videos [76.01906393673897]
We propose a self-supervised method to jointly learn 3D motion and depth from monocular videos.
Our system contains a depth estimation module to predict depth, and a new decomposed object-wise 3D motion (DO3D) estimation module to predict ego-motion and 3D object motion.
Our model delivers superior performance in all evaluated settings.
arXiv Detail & Related papers (2024-03-09T12:22:46Z) - DynaST: Dynamic Sparse Transformer for Exemplar-Guided Image Generation [56.514462874501675]
We propose a dynamic sparse attention based Transformer model to achieve fine-level matching with favorable efficiency.
The heart of our approach is a novel dynamic-attention unit, dedicated to covering the variation on the optimal number of tokens one position should focus on.
Experiments on three applications, pose-guided person image generation, edge-based face synthesis, and undistorted image style transfer, demonstrate that DynaST achieves superior performance in local details.
arXiv Detail & Related papers (2022-07-13T11:12:03Z) - Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection
Consistency [114.02182755620784]
We present an end-to-end joint training framework that explicitly models 6-DoF motion of multiple dynamic objects, ego-motion and depth in a monocular camera setup without supervision.
Our framework is shown to outperform the state-of-the-art depth and motion estimation methods.
arXiv Detail & Related papers (2021-02-04T14:26:42Z) - Learning to Segment Rigid Motions from Two Frames [72.14906744113125]
We propose a modular network, motivated by a geometric analysis of what independent object motions can be recovered from an egomotion field.
It takes two consecutive frames as input and predicts segmentation masks for the background and multiple rigidly moving objects, which are then parameterized by 3D rigid transformations.
Our method achieves state-of-the-art performance for rigid motion segmentation on KITTI and Sintel.
arXiv Detail & Related papers (2021-01-11T04:20:30Z) - Self-Supervised Joint Learning Framework of Depth Estimation via
Implicit Cues [24.743099160992937]
We propose a novel self-supervised joint learning framework for depth estimation.
The proposed framework outperforms the state-of-the-art(SOTA) on KITTI and Make3D datasets.
arXiv Detail & Related papers (2020-06-17T13:56:59Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.