Goto

Collaborating Authors

 fantastic beast


Supplementary Materials: AttrSeg: Open-Vocabulary Semantic Segmentation via Attribute Decomposition-Aggregation

Neural Information Processing Systems

The series is directed by David Yates and distributed by Warner Bros. It consists of three fantasy films as of 2022: Fantastic Beasts and Where to Find Them (2016) [1]. The movie follows Newt Scamander, a magizoologist who travels to New York with a suitcase full of magical creatures. When some of the creatures escape, he teams up with a group of people to find them before they cause any harm.


AttrSeg: Open-Vocabulary Semantic Segmentation via Attribute Decomposition-Aggregation

Neural Information Processing Systems

Open-vocabulary semantic segmentation is a challenging task that requires segmenting novel object categories at inference time. Recent works explore vision-language pre-training to handle this task, but suffer from unrealistic assumptions in practical scenarios, i.e., low-quality textual category names. For example, this paradigm assumes that new textual categories will be accurately and completely provided, and exist in lexicons during pre-training. However, exceptions often happen when meet with ambiguity for brief or incomplete names, new words that are not present in the pre-trained lexicons, and difficult-to-describe categories for users. To address these issues, this work proposes a novel attribute decomposition-aggregation framework, AttrSeg, inspired by human cognition in understanding new concepts.




AttrSeg: Open-Vocabulary Semantic Segmentation via Attribute Decomposition-Aggregation

arXiv.org Artificial Intelligence

Open-vocabulary semantic segmentation is a challenging task that requires segmenting novel object categories at inference time. Recent studies have explored vision-language pre-training to handle this task, but suffer from unrealistic assumptions in practical scenarios, i.e., low-quality textual category names. For example, this paradigm assumes that new textual categories will be accurately and completely provided, and exist in lexicons during pre-training. However, exceptions often happen when encountering ambiguity for brief or incomplete names, new words that are not present in the pre-trained lexicons, and difficult-to-describe categories for users. To address these issues, this work proposes a novel attribute decomposition-aggregation framework, AttrSeg, inspired by human cognition in understanding new concepts. Specifically, in the decomposition stage, we decouple class names into diverse attribute descriptions to complement semantic contexts from multiple perspectives. Two attribute construction strategies are designed: using large language models for common categories, and involving manually labeling for human-invented categories. In the aggregation stage, we group diverse attributes into an integrated global description, to form a discriminative classifier that distinguishes the target object from others. One hierarchical aggregation architecture is further proposed to achieve multi-level aggregations, leveraging the meticulously designed clustering module. The final results are obtained by computing the similarity between aggregated attributes and images embeddings. To evaluate the effectiveness, we annotate three types of datasets with attribute descriptions, and conduct extensive experiments and ablation studies. The results show the superior performance of attribute decomposition-aggregation.


Syfy's cliched but watchable 'Incorporated' envisions a disturbingly familiar future

Los Angeles Times

In the new Syfy series "Incorporated," it is the year 2074 and global warming has had its way with the world, 90% of which is controlled by multinational corporations who war over "dwindling resources." Some would say that this is already the case. Premiering Wednesday, it is a sometimes clever, just as often clichรฉd mix of dystopian tropes, with performances ranging from nicely modulated to almost over the top, and some sly design that, along with some twisted PSAs, also accounts for most of the story's humor. It is quite watchable and nothing special. Science fiction, it has often been noted, is all about the present and, besides the full-bore climate disasters we're rehearsing now, there are references to Canada building a fence to keep the Americans out, a declaration that "the system was rigged" (meaning the insurance business, but still), and midterm elections to which only 22% of voters turn out.


Conjure spells on your phone to prepare for 'Fantastic Beasts'

Engadget

J.K. Rowling's prequel to the Harry Potter series is set to hit theaters next week and Google wants to help you prepare by turning your phone into a magic wand. On an Android device, you can cast "spells" with an "OK Google" voice command. After you alert your phone with the magic phrase, follow it up with a command like "lumos" and "nox" to turn the flashlight on and off. You can also use "silencio" to mute any sounds and notifications. Sure, it's rather simple, but it's a neat way for Potter fans to get ready for Fantastic Beasts and Where to Find Them ahead of next week's debut.