Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312640109> ?p ?o ?g. }
- W4312640109 abstract "Numerous advancements in deep learning can be attributed to the access to large-scale and well-annotated datasets. However, such a dataset is prohibitively expensive in 3D computer vision due to the substantial collection cost. To alleviate this issue, we propose a cost-effective method for automatically generating a large amount of 3D objects with annotations. In particular, we synthesize objects simply by assembling multiple random primitives. These objects are thus auto-annotated with part labels originating from primitives. This allows us to perform multi-task learning by combining the supervised segmentation with unsupervised reconstruction. Considering the large overhead of learning on the generated dataset, we further propose a dataset distillation strategy to remove redundant samples regarding a target dataset. We conduct extensive experiments for the downstream tasks of 3D object classification. The results indicate that our dataset, together with multitask pretraining on its annotations, achieves the best performance compared to other commonly used datasets. Further study suggests that our strategy can improve the model performance by pretraining and fine-tuning scheme, especially for the dataset with a small scale. In addition, pretraining with the proposed dataset distillation method can save 86% of the pretraining time with negligible performance degradation. We expect that our attempt provides a new data-centric perspective for training 3D deep models." @default.
- W4312640109 created "2023-01-05" @default.
- W4312640109 creator A5036631624 @default.
- W4312640109 creator A5060983834 @default.
- W4312640109 creator A5071656520 @default.
- W4312640109 creator A5077099558 @default.
- W4312640109 creator A5088037430 @default.
- W4312640109 date "2022-06-01" @default.
- W4312640109 modified "2023-09-29" @default.
- W4312640109 title "Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives" @default.
- W4312640109 cites W1644641054 @default.
- W4312640109 cites W1971148082 @default.
- W4312640109 cites W2005756025 @default.
- W4312640109 cites W2108598243 @default.
- W4312640109 cites W2161112350 @default.
- W4312640109 cites W2211722331 @default.
- W4312640109 cites W2553307952 @default.
- W4312640109 cites W2557269700 @default.
- W4312640109 cites W2563685048 @default.
- W4312640109 cites W2594519801 @default.
- W4312640109 cites W2728326942 @default.
- W4312640109 cites W2767228836 @default.
- W4312640109 cites W2796426482 @default.
- W4312640109 cites W2798791840 @default.
- W4312640109 cites W2891396148 @default.
- W4312640109 cites W2901893665 @default.
- W4312640109 cites W2948080074 @default.
- W4312640109 cites W2955697347 @default.
- W4312640109 cites W2959771705 @default.
- W4312640109 cites W2961368225 @default.
- W4312640109 cites W2963231572 @default.
- W4312640109 cites W2963369474 @default.
- W4312640109 cites W2963509914 @default.
- W4312640109 cites W2964266557 @default.
- W4312640109 cites W2964341242 @default.
- W4312640109 cites W2972931660 @default.
- W4312640109 cites W2981010689 @default.
- W4312640109 cites W2981440248 @default.
- W4312640109 cites W2986615800 @default.
- W4312640109 cites W2989684653 @default.
- W4312640109 cites W2991485494 @default.
- W4312640109 cites W3007809903 @default.
- W4312640109 cites W3009750677 @default.
- W4312640109 cites W3034459762 @default.
- W4312640109 cites W3034700465 @default.
- W4312640109 cites W3034917353 @default.
- W4312640109 cites W3035534438 @default.
- W4312640109 cites W3102456204 @default.
- W4312640109 cites W3103201847 @default.
- W4312640109 cites W3106655268 @default.
- W4312640109 cites W3118644257 @default.
- W4312640109 cites W3191697454 @default.
- W4312640109 cites W3202611145 @default.
- W4312640109 cites W3203755984 @default.
- W4312640109 cites W4231186292 @default.
- W4312640109 cites W4255045236 @default.
- W4312640109 cites W4293659930 @default.
- W4312640109 doi "https://doi.org/10.1109/cvpr52688.2022.01548" @default.
- W4312640109 hasPublicationYear "2022" @default.
- W4312640109 type Work @default.
- W4312640109 citedByCount "0" @default.
- W4312640109 crossrefType "proceedings-article" @default.
- W4312640109 hasAuthorship W4312640109A5036631624 @default.
- W4312640109 hasAuthorship W4312640109A5060983834 @default.
- W4312640109 hasAuthorship W4312640109A5071656520 @default.
- W4312640109 hasAuthorship W4312640109A5077099558 @default.
- W4312640109 hasAuthorship W4312640109A5088037430 @default.
- W4312640109 hasBestOaLocation W43126401092 @default.
- W4312640109 hasConcept C108583219 @default.
- W4312640109 hasConcept C111919701 @default.
- W4312640109 hasConcept C119857082 @default.
- W4312640109 hasConcept C121332964 @default.
- W4312640109 hasConcept C124101348 @default.
- W4312640109 hasConcept C134306372 @default.
- W4312640109 hasConcept C153180895 @default.
- W4312640109 hasConcept C154945302 @default.
- W4312640109 hasConcept C162324750 @default.
- W4312640109 hasConcept C187736073 @default.
- W4312640109 hasConcept C19768560 @default.
- W4312640109 hasConcept C2778755073 @default.
- W4312640109 hasConcept C2779960059 @default.
- W4312640109 hasConcept C2780451532 @default.
- W4312640109 hasConcept C2781238097 @default.
- W4312640109 hasConcept C33923547 @default.
- W4312640109 hasConcept C41008148 @default.
- W4312640109 hasConcept C62520636 @default.
- W4312640109 hasConcept C77618280 @default.
- W4312640109 hasConcept C89600930 @default.
- W4312640109 hasConceptScore W4312640109C108583219 @default.
- W4312640109 hasConceptScore W4312640109C111919701 @default.
- W4312640109 hasConceptScore W4312640109C119857082 @default.
- W4312640109 hasConceptScore W4312640109C121332964 @default.
- W4312640109 hasConceptScore W4312640109C124101348 @default.
- W4312640109 hasConceptScore W4312640109C134306372 @default.
- W4312640109 hasConceptScore W4312640109C153180895 @default.
- W4312640109 hasConceptScore W4312640109C154945302 @default.
- W4312640109 hasConceptScore W4312640109C162324750 @default.
- W4312640109 hasConceptScore W4312640109C187736073 @default.
- W4312640109 hasConceptScore W4312640109C19768560 @default.
- W4312640109 hasConceptScore W4312640109C2778755073 @default.