Generative Learning of Differentiable Object Models for Compositional Interpretation of Complex Scenes

• URL: http://arxiv.org/abs/2506.08191v1
• Date: Mon, 09 Jun 2025 20:06:42 GMT
• Title: Generative Learning of Differentiable Object Models for Compositional Interpretation of Complex Scenes
• Abstract summary: This study builds on the architecture of the Disentangler of Visual Priors (DVP)<n>The DVP learns to interpret scenes by decomposing the perceived objects into independent visual aspects of shape, size, orientation, and color appearance.
• Score: 0.20718016474717196
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This study builds on the architecture of the Disentangler of Visual Priors (DVP), a type of autoencoder that learns to interpret scenes by decomposing the perceived objects into independent visual aspects of shape, size, orientation, and color appearance. These aspects are expressed as latent parameters which control a differentiable renderer that performs image reconstruction, so that the model can be trained end-to-end with gradient using reconstruction loss. In this study, we extend the original DVP so that it can handle multiple objects in a scene. We also exploit the interpretability of its latent by using the decoder to sample additional training examples and devising alternative training modes that rely on loss functions defined not only in the image space, but also in the latent space. This significantly facilitates training, which is otherwise challenging due to the presence of extensive plateaus in the image-space reconstruction loss. To examine the performance of this approach, we propose a new benchmark featuring multiple 2D objects, which subsumes the previously proposed Multi-dSprites dataset while being more parameterizable. We compare the DVP extended in these ways with two baselines (MONet and LIVE) and demonstrate its superiority in terms of reconstruction quality and capacity to decompose overlapping objects. We also analyze the gradients induced by the considered loss functions, explain how they impact the efficacy of training, and discuss the limitations of differentiable rendering in autoencoders and the ways in which they can be addressed.

Related papers

• An Adaptive, Disentangled Representation for Multidimensional MRI Reconstruction [3.3803617991341053]
  We present a new approach for representing and reconstructing multidimensional magnetic resonance imaging (MRI) data.<n>Our method builds on a novel, learned feature-based image representation that disentangles different types of features, such as geometry and contrast, into distinct low-dimensional latent spaces.<n>New reconstruction formulations and algorithms were developed to integrate the learned representation with a zero-shot selfsupervised learning adaptation and subspace modeling.
  arXiv  Detail & Related papers  (2025-12-31T07:02:21Z)
• Detection Based Part-level Articulated Object Reconstruction from Single RGBD Image [52.11275397911693]
  We propose an end-to-end trainable, cross-category method for reconstructing multiple man-made articulated objects from a single RGBD image.<n>We depart from previous works that rely on learning instance-level latent space, focusing on man-made articulated objects with predefined part counts.<n>Our method successfully reconstructs variously structured multiple instances that previous works cannot handle, and outperforms prior works in shape reconstruction and kinematics estimation.
  arXiv  Detail & Related papers  (2025-04-04T05:08:04Z)
• Decompositional Neural Scene Reconstruction with Generative Diffusion Prior [64.71091831762214]
  Decompositional reconstruction of 3D scenes, with complete shapes and detailed texture, is intriguing for downstream applications.<n>Recent approaches incorporate semantic or geometric regularization to address this issue, but they suffer significant degradation in underconstrained areas.<n>We propose DP-Recon, which employs diffusion priors in the form of Score Distillation Sampling (SDS) to optimize the neural representation of each individual object under novel views.
  arXiv  Detail & Related papers  (2025-03-19T02:11:31Z)
• Dynamic Scene Understanding through Object-Centric Voxelization and Neural Rendering [57.895846642868904]
  We present a 3D generative model named DynaVol-S for dynamic scenes that enables object-centric learning.<n>voxelization infers per-object occupancy probabilities at individual spatial locations.<n>Our approach integrates 2D semantic features to create 3D semantic grids, representing the scene through multiple disentangled voxel grids.
  arXiv  Detail & Related papers  (2024-07-30T15:33:58Z)
• Robust Geometry-Preserving Depth Estimation Using Differentiable Rendering [93.94371335579321]
  We propose a learning framework that trains models to predict geometry-preserving depth without requiring extra data or annotations. Comprehensive experiments underscore our framework's superior generalization capabilities. Our innovative loss functions empower the model to autonomously recover domain-specific scale-and-shift coefficients.
  arXiv  Detail & Related papers  (2023-09-18T12:36:39Z)
• Parametric Depth Based Feature Representation Learning for Object Detection and Segmentation in Bird's Eye View [44.78243406441798]
  This paper focuses on leveraging geometry information, such as depth, to model such feature transformation. We first lift the 2D image features to the 3D space defined for the ego vehicle via a predicted parametric depth distribution for each pixel in each view. We then aggregate the 3D feature volume based on the 3D space occupancy derived from depth to the BEV frame.
  arXiv  Detail & Related papers  (2023-07-09T06:07:22Z)
• CbwLoss: Constrained Bidirectional Weighted Loss for Self-supervised Learning of Depth and Pose [13.581694284209885]
  Photometric differences are used to train neural networks for estimating depth and camera pose from unlabeled monocular videos. In this paper, we deal with moving objects and occlusions utilizing the difference of the flow fields and depth structure generated by affine transformation and view synthesis. We mitigate the effect of textureless regions on model optimization by measuring differences between features with more semantic and contextual information without adding networks.
  arXiv  Detail & Related papers  (2022-12-12T12:18:24Z)
• RISP: Rendering-Invariant State Predictor with Differentiable Simulation and Rendering for Cross-Domain Parameter Estimation [110.4255414234771]
  Existing solutions require massive training data or lack generalizability to unknown rendering configurations. We propose a novel approach that marries domain randomization and differentiable rendering gradients to address this problem. Our approach achieves significantly lower reconstruction errors and has better generalizability among unknown rendering configurations.
  arXiv  Detail & Related papers  (2022-05-11T17:59:51Z)
• Contrastive Object-level Pre-training with Spatial Noise Curriculum Learning [12.697842097171119]
  We present a curriculum learning mechanism that adaptively augments the generated regions, which allows the model to consistently acquire a useful learning signal. Our experiments show that our approach improves on the MoCo v2 baseline by a large margin on multiple object-level tasks when pre-training on multi-object scene image datasets.
  arXiv  Detail & Related papers  (2021-11-26T18:29:57Z)
• Unsupervised Learning of 3D Object Categories from Videos in the Wild [75.09720013151247]
  We focus on learning a model from multiple views of a large collection of object instances. We propose a new neural network design, called warp-conditioned ray embedding (WCR), which significantly improves reconstruction. Our evaluation demonstrates performance improvements over several deep monocular reconstruction baselines on existing benchmarks.
  arXiv  Detail & Related papers  (2021-03-30T17:57:01Z)

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.