Inter3D: A Benchmark and Strong Baseline for Human-Interactive 3D Object Reconstruction
Authors: Gan Chen, Ying He, Mulin Yu, F.Richard Yu, Gang Xu, Fei Ma, Ming Li, Guang Zhou
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct extensive experiments on the proposed benchmark, showcasing the challenges of the task and the superiority of our approach. The code and data are publicly available at https://github.com/Inter3D-ui/ Inter3D. |
| Researcher Affiliation | Collaboration | Gan Chen1 , Ying He1,2 , Mulin Yu3 , F. Richard Yu1,2 , Gang Xu2 , Fei Ma2 , Ming Li2 and Guang Zhou2 1College of Computer Science and Software Engineering, Shenzhen University 2Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) 3Shanghai Artificial Intelligence Laboratory EMAIL, EMAIL, EMAIL, EMAIL, EMAIL, EMAIL |
| Pseudocode | No | The paper describes the approach in prose and figures, such as Figure 2 'Overview of our approach', but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code and data are publicly available at https://github.com/Inter3D-ui/ Inter3D. |
| Open Datasets | Yes | We introduce a self-collected dataset featuring commonly encountered interactive objects and a new evaluation pipeline... The code and data are publicly available at https://github.com/Inter3D-ui/ Inter3D. |
| Dataset Splits | Yes | For each object, 60 images are extracted through uniform sampling at regular intervals to represent the canonical state and each individual state. ... only the canonical and individual part states of an object are observed during training, while all combination states remain unseen and are reserved for evaluation... |
| Hardware Specification | Yes | All experiments are conducted on a single NVIDIA Ge Force RTX 3090 GPU. |
| Software Dependencies | No | The paper mentions using the Adam optimizer but does not specify its version or any other software dependencies with version numbers (e.g., Python, PyTorch, CUDA versions). |
| Experiment Setup | Yes | We employ the Adam optimizer [Kingma and Ba, 2014] with hyperparameters lr = 1e-2, β1 = 0.9, β2 = 0.99, and ϵ =1e15 to train our model. The image resolution in our dataset is 1280 × 680. Each epoch consists of 1,000 iterations and 8,192 pixels are randomly selected for each iteration. The Canonical Modelling stage requires two epochs to reconstruct the canonical shape of an interactive object. During the Movable Part Decomposition stage, a random individual state is selected for each iteration to learn the corresponding movable part. This phase takes 13 epochs. ... The learning objective in Canonical Modelling LCM is as follows: LCM = λ1Ldist + λ2Lopacity + LMSE, (9) where λ1 and λ2 are the weighting coefficients, both set to 1e-3 in our experiments. In Movable Part Decomposition... LMP D = λ3Lconsis + λ4Lsmooth + λ5LMSR, (10) where λ3, λ4 and λ5 are set as 1e-4, 1e-3, and 1e-2, respectively. |