1. Home
  2. Detail

@shimaamahmoud1312: #shimaa_mahmoud

shimaa_شيماء 🧑‍🍳🧑‍🍳♥️🥰😋
shimaa_شيماء 🧑‍🍳🧑‍🍳♥️🥰😋
Open In TikTok:
Region: EG
Monday 21 September 2026 20:17:34 GMT
193
13
2
1

Music

Download

No Watermark .mp4 (2.42MB) No Watermark(HD) .mp4 (2.42MB) Watermark .mp4 (0MB) Music .mp3

Comments

moag_395
مطبخ البيت :
🥰🥰🥰🥰🥰🥰
2026-09-22 16:54:27
0
tofeektofeektofee0
فتى اليمن :
[مؤثر][مؤثر][مؤثر]
2026-09-23 03:44:04
0
To see more videos from user @shimaamahmoud1312, please go to the Tikwm homepage.

Other Videos

#พี่ชายสองคนชื่ออะไรกันจ๊ะ😹
#พี่ชายสองคนชื่ออะไรกันจ๊ะ😹
✌️#tâmtrạng
✌️#tâmtrạng
الجنزير التقيل AI | أغنية تراثية من غرب السودان
الجنزير التقيل AI | أغنية تراثية من غرب السودان "مردوم " تراث المردوم بصنفو الفلكوري في السودان .. لما فيهو من تميز لجمال التلحينه والكلمات الدراجيه للكنه قبائل كردفان، جزء من طريقة الاداء مستنسخ من الأغنية للجزار وميادة #sudanese_ai_songs #السودان_مشاهير_تيك_توك #sudanese_tiktokمشاهير🇸🇩
Я опять пошёл, а вы тут разбирайтесь 🤝👽 #песни #новинкимузыки #плейлист #музыкадлядуши🎶🎧 #треки
Я опять пошёл, а вы тут разбирайтесь 🤝👽 #песни #новинкимузыки #плейлист #музыкадлядуши🎶🎧 #треки
Cutting a trailer means watching a two hour film and remembering which thirty seconds carry an emotion. It runs on memory and taste, it takes many passes, and the first real audience reaction arrives after the marketing is already out. A media client wanted a model that reads the whole film and marks the candidate moments. The brief said sentiment, but one sentiment score per scene tells an editor almost nothing. ⠀ They are hunting a specific beat. The character is alone, the line lands, the music turns, the shot is wide enough to cut on. So I widened the analysis to tag characters, actions, dialogue, mood, objects, setting, plot, pace, colour, humour and audio effects, every one of them against a timecode. The output schema was the product. ⠀ A section of film becomes a block with a start time, an end time, and the same set of fields every time, which makes the output queryable. An editor can ask for every block where a named character is grieving and the pace is slow. I also segmented before analysing, splitting the film into acts and then into blocks of a few seconds to about a minute, each analysed with the surrounding context supplied alongside it. What is sitting in your archive that nobody has time to watch? ⠀ Follow @gamechangerai for more. ⠀ #ai #machinelearning #computervision #multimodalai #videoediting
Cutting a trailer means watching a two hour film and remembering which thirty seconds carry an emotion. It runs on memory and taste, it takes many passes, and the first real audience reaction arrives after the marketing is already out. A media client wanted a model that reads the whole film and marks the candidate moments. The brief said sentiment, but one sentiment score per scene tells an editor almost nothing. ⠀ They are hunting a specific beat. The character is alone, the line lands, the music turns, the shot is wide enough to cut on. So I widened the analysis to tag characters, actions, dialogue, mood, objects, setting, plot, pace, colour, humour and audio effects, every one of them against a timecode. The output schema was the product. ⠀ A section of film becomes a block with a start time, an end time, and the same set of fields every time, which makes the output queryable. An editor can ask for every block where a named character is grieving and the pace is slow. I also segmented before analysing, splitting the film into acts and then into blocks of a few seconds to about a minute, each analysed with the surrounding context supplied alongside it. What is sitting in your archive that nobody has time to watch? ⠀ Follow @gamechangerai for more. ⠀ #ai #machinelearning #computervision #multimodalai #videoediting


About

Legal