iCaps: Iterative Category-level Object Pose and Shape Estimation

Dieter Fox; Junyi Geng; Timothy Bretl; Xinke Deng; Yu Xiang

RECORD

iCaps: Iterative Category-level Object Pose and Shape Estimation

Xinke Deng, Junyi Geng, Timothy Bretl, Yu Xiang, Dieter Fox

Computer Vision and Pattern Recognition · Submitted 31 Dec 2021

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim We evaluate our approach on a publicly available dataset and demonstrate its effectiveness.

how it is stated · asserted 1.00 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2201.00059 (version 1)

Pith IDQCWAMOPP, Pith's permanent record of this paper and its review

CitationDieter Fox, Junyi Geng, Timothy Bretl, Xinke Deng, Yu Xiang. "iCaps: Iterative Category-level Object Pose and Shape Estimation." arXiv:2201.00059 (2021). https://pith.science/paper/QCWAMOPP

Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 6,141 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.20 · human participants 0.02 · controlled comparison 0.06 · randomized assignment 0.01 · states quantitative result 0.05 · compares with prior results 0.12 · mentions code or data availability 0.04.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

This paper proposes a category-level 6D object pose and shape estimation approach iCaps, which allows tracking 6D poses of unseen objects in a category and estimating their 3D shapes. We develop a category-level auto-encoder network using depth images as input, where feature embeddings from the auto-encoder encode poses of objects in a category. The auto-encoder can be used in a particle filter framework to estimate and track 6D poses of objects in a category. By exploiting an implicit shape representation based on signed distance functions, we build a LatentNet to estimate a latent representation of the 3D shape given the estimated pose of an object. Then the estimated pose and shape can be used to update each other in an iterative way. Our category-level 6D object pose and shape estimation pipeline only requires 2D detection and segmentation for initialization. We evaluate our approach on a publicly available dataset and demonstrate its effectiveness. In particular, our method achieves comparably high accuracy on shape estimation.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	This paper proposes a category-level 6D object pose and shape estimation approach iCaps, which allows tracking 6D poses of unseen objects in a category and estimating their 3D shapes.	contribution 1.00	0.86	0.35	asserted 1.00
s1	We develop a category-level auto-encoder network using depth images as input, where feature embeddings from the auto-encoder encode poses of objects in a category.	method 0.92	0.94	0.05	asserted 1.00
s2	The auto-encoder can be used in a particle filter framework to estimate and track 6D poses of objects in a category.	method 0.98	0.71	0.05	asserted 0.58
s3	By exploiting an implicit shape representation based on signed distance functions, we build a LatentNet to estimate a latent representation of the 3D shape given the estimated pose of an object.	method 0.94	0.93	0.04	asserted 1.00
s4	Then the estimated pose and shape can be used to update each other in an iterative way.	method 0.98	0.83	0.07	asserted 0.99
s5	Our category-level 6D object pose and shape estimation pipeline only requires 2D detection and segmentation for initialization.	method 0.84	0.71	0.05	asserted 1.00
s6	We evaluate our approach on a publicly available dataset and demonstrate its effectiveness.	result 0.70	0.96	0.27	asserted 1.00
s7	In particular, our method achieves comparably high accuracy on shape estimation.	result 0.99	0.96	0.25	asserted 1.00

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.86.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper

Original abstract

This paper proposes a category-level 6D object pose and shape estimation approach iCaps, which allows tracking 6D poses of unseen objects in a category and estimating their 3D shapes. We develop a category-level auto-encoder network using depth images as input, where feature embeddings from the auto-encoder encode poses of objects in a category. The auto-encoder can be used in a particle filter framework to estimate and track 6D poses of objects in a category. By exploiting an implicit shape representation based on signed distance functions, we build a LatentNet to estimate a latent representation of the 3D shape given the estimated pose of an object. Then the estimated pose and shape can be used to update each other in an iterative way. Our category-level 6D object pose and shape estimation pipeline only requires 2D detection and segmentation for initialization. We evaluate our approach on a publicly available dataset and demonstrate its effectiveness. In particular, our method achieves comparably high accuracy on shape estimation.

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

What this paper does with the arXiv papers it cites

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

What it cites · 3 sentences without a confident label

Uses data or values from 1 paper

We take advantage of ShapeNet [ref] , which is a large scale database of 3D CAD models.

ShapeNet: An Information-Rich 3D Model Repository · Category-level Auto-encoder · 0.97

Compares with 2 papers

We compare our category-level 6D object pose tracking system with ICP [ref] , 6-PACK [ref] , CAPTRA [ref] , and BundleTrack [ref] .

Open3D: A Modern Library for 3D Data Processing · Results · 0.98
We compare our category-level 6D object pose tracking system with ICP [ref] , 6-PACK [ref] , CAPTRA [ref] , and BundleTrack [ref] .

CAPTRA: CAtegory-level Pose Tracking for Rigid and Articulated Objects from Point Clouds · Results · 0.98

Cites as background 2 papers

Besides the mentioned work on category-level pose estimation for single frames such as NOCS [ref] , CASS [ref] , FS-Net [ref] and the category-level tracking work 6-PACK [ref] , recently, Weng et al. [ref] propose a unified framework that can handle 9DoF category-level pose tracking for rigid object instances.

CAPTRA: CAtegory-level Pose Tracking for Rigid and Articulated Objects from Point Clouds · Related Work · 1.00
PoseCNN [ref] designs an end-to-end network for 6D object pose estimation based on the VGG architecture [ref] .

Very Deep Convolutional Networks for Large-Scale Image Recognition · Related Work · 0.90

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.